Part I. Scandium Hydride and Alkyl Complexes with a Linked Monocyclopentadienyl-Amido Ligand-Framework: Single Component Catalysts for the Polymerization of α-Olefins. Part II. Hydrazido(1-) and 2,2-Dimethylhydrazido(1-) Derivatives of Permethylscandocene. Preparation and Structural Characterization of their Products from Reactions with Acetonitrile Citation Shapiro, Pamela Joy (1991) Part I. Scandium Hydride and Alkyl Complexes with a Linked Monocyclopentadienyl-Amido Ligand-Framework: Single Component Catalysts for the Polymerization of α-Olefins. Part II. Hydrazido(1-) and 2,2-Dimethylhydrazido(1-) Derivatives of Permethylscandocene. Preparation and Structural Characterization of their Products from Reactions with Acetonitrile. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/hmc3-gy25. https://resolver.caltech.edu/CaltechTHESIS:11042019-173308161 Abstract Various monocyclopentadienyl scandium amido derivatives with the general formulas {(η 5 -C 5 Me 4 )SiMe 2 (η 1 -NCMe 3 }ScR, (Cp*SiNR)ScR, and {(η 5 -C 5 H 3 CMe 3 )SiMe 2 (η 1 -NCMe 3 }ScR, ( tBu CpSiNR)ScR, have been prepared. [(Cp*SiNR)(PMe 3 )ScH] 2 is prepared from the hydrogenation of (Cp*SiNR)ScCH(SiMe 3 ) 2 in the presence of PMe 3 . The scandium hydride effects the multiple insertion of α-olefins in a controlled fashion to form low molecular weight polymers. β-H elimination appears to be a principal chain-transfer pathway in this catalyst system. The stoichiometric reaction between [(Cp*SINR)(PMe 3 )ScH] 2 and two equivalents of ethylene produces the ethylene bridged dimer [(Cp*SiNR)(PMe 3 )Sc] 2 (µ, η 2 , η 2 -C 2 H 4 ) and an equivalent of ethane. The stoichiometric reaction between the scandium hydride and either propene or butene affords the PMe 3 -free, alkyl-bridged scandium dimers [(Cp*SiNR)Sc] 2 (µ-CH 2 CH 2 CH 3 ) 2 and [(Cp*SiNR)Sc] 2 (µ-CH 2 CH 2 CH 2 CH 3 ) 2 . The absence of coordinating phosphine makes these complexes more active olefin polymerization catalysts. The complexes (Cp*SiNR)(PMe 3 )ScCH 2 CH(CH) 3 CH 2 CH 3 and (Cp*SiNR)(PMe 3 )ScCH(C 6 H 5 )CH 2 CH 2 CH 2 (C 6 H 5 ) have also been prepared and spectroscopically characterized. Kinetic analysis of the rate dependence of 1-pentene polymerization by [(Cp*SiNR)(PMe 3 )ScH] 2 on added PMe 3 reveals a phosphine dissociation pre-equilibrium to form a PMe 3 -free active intermediate. The monomeric nature of this intermediate is revealed by the kinetic analysis of 1-pentene polymerization by [(Cp*SiNR)Sc] 2 (µ-CH 2 CH 2 CH 3 ) 2 as a function of scandium concentration. (Cp*SiNR)ScR was found to be >99% head-to-tail regioselectfve in the polymerization of propene. Preliminary 13 C NMR data indicate lower regioselectivity in propene polymerization by ( tBu CpSiNR)ScR. A slight syndiotactic preference is observed in the stereochemistry of the polypropene produced with [(Cp*SiNR)Sc] 2 (µ-CH 2 CH 2 CH 3 ) 2 . By contrast, [( tBu CpSiNR)ScMe] x displays no stereocontrol, producing an atactic polymer. A rare example of an unsubstituted hydrazido(1-) complex, Cp* 2 ScNHNH 2 (Cp* = (η 5 -C 5 Me 5 )), prepared by reaction of one equivalent of anhydrous hydrazine with Cp* 2 ScCH 3 , reacts with acetonitrile to form [chemical formula; see abstract in scanned thesis for details]. The crystal structure of [chemical formula; see abstract in scanned thesis for details] reveals a five-membered, nearly planar [chemical formula; see abstract in scanned thesis for details] ring. The results of a labeling study using 15 N=CCH 3 are consistent with a mechanism involving insertion of acetonitrile into the Sc-N bond of Cp* 2 ScNHNH 2 , followed by tautomerization to form [chemical formula; see abstract in scanned thesis for details]. The closely related compounds Cp* 2 ScNHNMe 2 and [chemical formula; see abstract in scanned thesis for details] have also been prepared, and the structure of the latter compound has been determined. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: Chemistry Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemistry Thesis Availability: Public (worldwide access) Research Advisor(s): Bercaw, John E. Thesis Committee: Gray, Harry B. (chair) Bercaw, John E. Roberts, John D. Grubbs, Robert H. Defense Date: 19 June 1990 Other Numbering System: Other Numbering System Name Other Numbering System ID UMI 9113620 Record Number: CaltechTHESIS:11042019-173308161 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:11042019-173308161 DOI: 10.7907/hmc3-gy25 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 11892 Collection: CaltechTHESIS Deposited By: Mel Ray Deposited On: 05 Nov 2019 17:47 Last Modified: 16 Apr 2021 23:06 Thesis Files Preview PDF - Final Version See Usage Policy. 5MB Repository Staff Only: item control page

Part I. Scandium H ydride and Alkyl Com plexes w ith A Linked M onocyclopentadienyl-Am ido Ligand-Fram ework: Single Com ponent Catalysts fo r the Polym erization of a-O lefins Part II. H yd razid o (l-) and 2,2-D im ethylhydrazido(1-) Derivatives of Perm ethylscandocene. Preparation and Structural Characterization of Their Products from Reactions w ith Acetonitrile Thesis by Pam ela J. Shapiro In Partial Fulfillm ent of the R equirem ents fo r the Degree of Doctor o f Philosophy California Institute of Technology Pasadena, California 1991 (subm itted June 1 9 ,1 9 9 0 ) To My Grandmother ACKNOWLEDGEMENTS Believe it or not, I’ve learned several lessons about life over these past five years at Caltech in addition to my teaming about chemistry. The people mentioned below have contributed to my growth in one or both of these areas: General thanks go to the Bercaw group for providing a stimulating and entertaining environment in which to learn. Special thanks go to John Bercaw for his patient teachings and his infectious enthusiasm for science. I am grateful to Emilio Bunel for teaching me the tricks of the trad e and for making me feel guilty about not working hard enough. Trim initiated me to vacuum line techniques and w as a patient line-mate. Barb introduced me to "swivel-frit" technology. I don’t know how I would have gotten anything done without th e services of people such as John Pirolo, Gabor Faludi, Tom Dunn, Larry Henling and Fenton Harvey. Scott Ross, Dave W heeler and Dom McGrath are gratefully acknowledged for assistance with use of the high field NMR instruments. Bill Schaefer is gratefully acknowledged for performing most of th e crystal structure determinations in this thesis. Thanks go to Pat Anderson for fashion and diet tips, rice cakes and pleasant conversation. Thanks go to Andrew H. for being my aerobics buddy, for his cheerful "hello’’ and for instruction on the 500 MHz NMR. Thank you, Leroy, for letting me graduate 25th and for bringing in a Hanuka bush every December. John P., Donnie, Janet, Bryan, Eric, Warren, Helmut and Kaspar are acknowledged for their encouragement, support and teacnings about chemistry. I am grateful to Harry Gray for always making me fee l important Many thanks go to Jay Labinger for his receptiveness to questions and for extensive help with the polymerization kinetics discussed in Chapter 2. Mark Olson is the Gauss expert who helped me with the non linear least-squares analyses discussed in Chapter 2. I will be eternally grateful to Roger for his generosity, reliability and willingness to accompany me on frozen yogurt and Winchell's-muffin breaks. Thanks go to David Montgomery for being a very special friend over this past year. He may think that he came to Caltech for his benefit, but I am convinced he was sent here for mine. Most importantly, I am grateful to my grandmother for her constant love, support and encouragement. V ABSTRACT Various monocyclopentadienyl scandium amido derivatives with the general formulas {(*;5-C 5 Me4 )SiMe2 (»j1-NCMe3 }ScR, (Cp*SiNR)ScR, and {(»J5-C 5 H3 CMe3 )SiMe 2 (»71-NCMe3 }ScR, (tBuCpSiNR)ScR, have been prepared. [(Cp*SiNR)(PMe 3 )ScH ] 2 is prepared from the hydrogenation of (Cp*SiNR)ScCH(SiMe3 ) 2 in the presence of PMe3 . The scandium hydride effects the multiple insertion of a-olefins in a controlled fashion to form low molecular weight polymers. flM elimination appears to be a principal chain-transfer pathway in this catalyst system. The stoichiometric reaction between [(Cp*SINR)(PMe3)ScH ] 2 and two equivalents of ethylene produces the ethylene bridged dimer [(Cp*SiNR)(PMe3)Sc]2 (p. n2. f?2-C 2 H4 ) and an equivalent of ethane. The stoichiometric reaction between the scandium hydride and either propene or butene affords the PMe3 *free, alkyl-bridged scandium dimers [(Cp*SiNR)Sc]2 (p-CH 2 CH 2 CH 3 ) 2 and [(Cp*SiNR)Sc]2 (M*CH2 CH 2 CH 2 CH 3 )2 . The absence of coordinating phosphine makes these complexes more active olefin polymerization catalysts. The complexes (Cp*SiNR)(PMe3 )ScCH 2 CH(CH)3CH2 CH3 and (Cp*SiNR)(PMe3 )ScCH(CeHs)CH 2 CH 2 CH 2 (CgHs) have also been prepared and spectroscopically characterized. Kinetic analysis of the rate dependence of 1-pentene polymerization by [(Cp*SiNR)(PMe 3 )ScH ] 2 on added PMea reveals a phosphine dissociation pre-equilibrium to form a PMe3 -free active intermediate. The monomeric nature of this intermediate is revealed by the kinetic analysis of 1-pentene polymerization by [(Cp*SINR)Sc]2 0 *-CH 2 CH2 CH 3 ) 2 as a function of scandium concentration. (Cp*SiNR)ScR was found to be >99% head-to-tail regloselectfve in the polymerization of propene. Preliminary 13C NMR data indicate lower regioselectivity in propene polymerization by (tBuCpSINR)ScR. A slight syndiotactic preference is observed in the stereochemistry of the polypropene produced with [(Cp*SiNR)Sc]2 (p-CH2CH 2 CH 3 )2 . By contrast, [(tBuCpSiNR)ScMe]x displays no stereocontrol, producing an atactic polymer. A rare example of an unsubstituted hydrazido(l-) complex, Cp*2 ScNHNH2 (Cp* = (r;5-CsMe5 )), prepared by reaction of one equivalent of anhydrous hydrazine with Cp*2ScCH 3 , reacts with acetonitrile to form C p*2 ScN(H)C(CH 3 )Nf\IH2. The crystal structure of Cp*2ScN(H)C(CH 3 )NfvlH2 reveals a five-membered, nearly planar [Sc-NH-C(CH 3 )-N-tvJH2 ] ring. The results of a labeling study using 15N«CCH3 are consistent with a mechanism involving insertion of acetonitrile into the Sc-N bond of Cp*2 ScNHNH 2 , followed by tautomerization to form Cp*2 ScN (H)C(CH 3 ) NfslH2 - The closely related compounds I------------------- 1 Cp*2 ScNHNMe2 and Cp*2 ScN(H)C(CH 3 )NNMe2 have also been prepared, and the structure of the latter compound has been determined. TABLE OF CONTENTS Acknowledgements Abstract Part I: Scandium Hydride and Alkyl Complexes with A Linked Monocyclopentadienyl-Amido Ligand-Framework: Single Component Catalysts for the Polymerization of a-Olefins Chapter 1 The Synthesis and Structural Characterization of (Cp*SiNR)ScR and (tBuCpSiNR)ScR Complexes Chapter 2 Investigations of the a-Olefin Polymerization Adivity and Mechanism for Monocyclopentadienyl Scandium Amido Complexes Part II: Hydrazido(l-) and 2,2-Oimethylhydrazido(1-) Derivatives of Permethylscandocene. Preparation and Structual Characterization of Their Products from Reactions with Acetonitrile 1 Parti Scandium Hydride and Alkyl Complexea with A Linked Monocyclopentadlenyl-Amido Ligand-Framework: Single Component Catalysts for the Polymerization of a-OlefIns. 2 INTRODUCTION In 1953 Ziegler at al. made the serendipitous discovery that the binary system of a transition metal salt with an alkyl-aluminum species catalyzes the polymerization of ethylene to linear, high molecular weight, high density polyethylene.1 Shortly thereafter, Natta and coworkers applied these catalysts to the polymerization of propene and other a-olefins and demonstrated the ability of these systems to produce stereoregular (isotactic) macromolecules.2 These discoveries coupled with the independent discovery of ethylene polymerization catalysis by metal oxides supported on silica or alumina, most notably the Phillips Petroleum CrO3/Si02 catalyst,3 virtually revolutionized the polymer industry. Although new generations of Zlegler-Natta catalyst systems which are more active and more stereospecific have developed at a rapid pace, the structures of the active catalysts have remained elusive.4 As with many catalysts, direct characterization of these systems has been complicated by their heterogeneous, multicomponent composition and the low concentration of active centers in the bulk material. Models for the active centers have relied on indirect studies of polymerization kinetics, polymer microstructure and molecular weight distribution and acthre-site concentration. As a result the structures for the active catalysts have been highly speculative.3 The most favored mechanistic model is that developed by Cossee and Ariman.® The primary feature of this model is the insertion of the olefin into the M-C bond of a transition metal alkyl species (Scheme 1). R X /X M X R R R X X /X -M; X Scheme 1. Cossee-Ariman Mechanlem for Olefin Polymerization X 3 Although this mechanism was proposed for an octahedral titanium site in a heterogenous TiCb-based catalyst system, compelling evidence in support of it has recently been derived from soluble, well-defined organometallic model systems.7 An alternative mechanism put forth by Green and coworkers also involves propagation at the transition metal but requires the formation of a metal carbene intermediate, which adds olefin in a 2+2 fashion, as in the olefin metathesis reaction (Scheme 2).® C H C c M, C etc. C Scheme 2. Gresn-Rooney Mechanism for Olefin Polymerization So far, only one well-defined, ethylene-polymerizing organometallic species, a tantalum alkylidene complex reported by Turner and Schrock, has appeared to operate by the Green-Rooney mechanism.® Most other organometallic Ziegler-Natta model systems are d° and therefore cannot undergo the two electron oxidation state change associated with a-H abstraction. A more subtle form of C-H activation involving an agostic interaction for assisting olefin insertion has been proposed for these d° systems (Scheme 3).10 The involvement of an a-agostic interaction in the olefin-insertion mechanism is currently being explored by our group as well as by others.11 Schema 3. Agostic a-C-H Assisted Mechanism for Olafln Polymerization 4 It is through these well-defined, single-component organometallic systems that much of our understanding of elementary processes pertinent to Ziegler-Natta polymerization, such as olefin insertion and the microscopic reverse reactions of /8-H and /9-alkyl elimination, has evolved. In fact, many of these systems, based mainly on eariy transition metals or lanthanide metals (Figure 1), are quite active catalysts for the polymerization of ethylene. Their utility in unravelling the mechanism of Ziegler-Natta catalysis, especially with respect to the remarkable regio- and stereospecificity of most Ziegler-Natta systems, has been limited, however, by their inability to polymerize a-olefins. At best, a few of these systems have been reported to oligomerize propene to chain lengths of C o ,.™ '* Reported herein are well-defined organoscandium systems that are rare examples of single-component, homogeneous catalysts for the polymerization of a-olefins. The systems, (Cp*SINR)ScR and (tBuCpSINR)ScR (see below), which will be described may be considered the third generation of organoscandium compounds that have emerged from our research group. These complexes differ from their predecessors, (r}S-CsMes)2 ScR,12 {(r75 ^ 5Me4)2SiMe2}ScR and { (»>5-C 5 H 3 CMe3 )2 SiMe2 }ScR,13 in that one of the cyclopentadienyl ligands on the scandium is replaced by a bulky amido group. "(Cp'SiNR)ScR" "(^CpSiNRJScR" 5 CH. ’CH. BPh4 Sc— R THF BPh* THF M-Nd. Sm, Lu Figure 1. Some Organometallic Systems that Polymariaa Ethylana 6 The reactivity of the scandium has been found to be highly sensitive to changes in the sterics of its supporting ligand system. Thus, we were guided by the reactivity profiles of the earlier organoscandium systems in choosing this ligand modification as a logical next step in our efforts to design a catalyst system for the multiple insertion of a-olefins.14 The syntheses of various derivatives of these monocyclopentadienyl scandium amido complexes, along with their spectroscopic and x-ray crystallographic characterization, is discussed in the first chapter of this thesis. Characterization of the polymerization activity of these systems and experiments to elucidate the nature of the active catalyst are covered in Chapter 2. 7 R eferences 1. Ziegler, K.; Holzkamp, E.; Breil, H; Martin, H. Angew.Chem. 1955,67, 541. 2. Natta, G.; Pino, P.; Corradini, P.; Danusso, F.; Mantica, E.; Mazzanti, G.; Moraglio, G. J. Am. Chem. Soc. 1955, 77,1708. 3. Clark, A.; Hogan, J. P.; Banks, R. L ; Lanning, W.C. Ind. Eng. Chem., 1 9 5 6 ,4 8 ,1152. 4. For reviews on Ziegler-Natta polymerization, see: (a) Sinn, H.; Kaminsky, W. Adv. Organomet. Chem. 1 9 8 0 ,99. (b) Pino, P.; Mulhaupt, R. Angew. Chem., Int. Ed. Engl. 1 9 8 0 ,1 9 ,857. (c) Boor, J. Ziegler-Natta Catalysts and Polymerizations; Academic Press: New York, 1979. (d) Tait, P. J. T. In Comprehensive Polymer Science; Allen, G.; Bevington, J. C., Eds.; Pergamon Press: Oxford, 1989; Chapter 1. 5. Tait, P. J. T. In Comprehensive Polymer Science; Allen, G.; Bevington, J. C., Eds.; Pergamon Press: Oxford, 1989; Chapter 2. 6. a) Cossee, P. J. Catal. 1 9 6 4 ,3 ,80. b) Arlman, E. J.; Cossee, P. J. Catal. 1 9 6 4 ,3 ,9 9 . 7. (a) Watson, P. L; Parshall, G. W. Acc. Chem. Res. 1985,18,51 and references therein, (b) Watson, P. L; Herskovitz, T. ACS Symp. Ser. 1983, No. 212,459. (c) Jordon, R. F. J. Chem. Educ. 1988,65(4), 285 and references therein, (d) Jeske, G.; Lauke, H.; Mauermann, H.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1 9 8 5 ,107,8091. (e) Jeske, G.; Schock, L E.; Swepston, P. N.; Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1 9 8 5 ,107,8103. (0 Ballard, D. G. H.; Courtis, A.; Holton, J.; McMeeking, J.; Pearce, R. J. Chem. Soc* Chem. Commun. 1978,994. (g) Schmidt, G. F.; Brookhart, M. J.Am . Chem. Soc. 1 9 8 5 ,1 0 7 ,1443. (h) Thomas, B. J.; Theopold, K. H. J. Am. Chem. Soc. 1 9 8 8 ,1 1 0 ,5902. (I) Taube, R.; Krakowa, L J. Organomet Chem. 1 9 8 8 ,347, C9. 0) Hlatky, G. G.; Turner, H. W.; Eckman, R. R. J. Am. Chem. Soc. 1 9 8 9 ,111,2728. 8. Ivin, K. J.; Rooney, J. J.; Stewart, C. 0.; Green, M. L H.; Mahtab, R. J. Chem. Soc. Chem. Commun. 1978,804. 9. Turner, H. W.; Schrock, R. R. J. Am. Chem. Soc. 1 9 8 2 ,1 04,2331. 10. Brookhart, M.; Green, M. L H. J. Organometal. Chem. 1 9 8 3 ,2 5 0 ,395. 11. a) An interaction of the hydrogen on the a-carbon with either the transition metal or the aluminum in a Ziegler-Natta catalyst system has been probed by examinination of the deuterium isotope effect on the stereochemistry of intramolecular olefin insertion by titanocene alkenyl chloride complexes catalyzed by ethylaluminum dichloride (Clawson, L ; Soto, J.; Buchwald, S. L ; Stelgerwald, M. L ; Grubbs, R. H. J. Am. Chem. Soc. 1 9 8 5 ,107,3377). Similar experiments involving the cyclization of deuterium labeled a,w-diolefins by Op(PMes)ScH are currently being performed in our labs (Piers, W. E., unpublished results), b) Brintzinger and coworkers have calculated that an a-C-H agostic interaction stabilizes the transition state for olefin insertion by ca 20 kcal mol*1 (personal communication). 8 12. Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C.; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. J. Am. Chem. Soc. 1 9 8 7 ,109,203. 13. Bunel, E.; Burger, B. J.; Bercaw, J. E. J.Am . Chem. Soc. 1 9 8 8 ,110, 976. 14. For a review of our work with organoscandium compounds see: Piers, W. E.; Shapiro, P. J.; Bunel, E. E.; Bercaw, J. E. Synlett, 1 9 9 0 ,2 , 74. Chapter 1 The Synthesis and Structural Characterization of (Cp*SINR)ScR and (tBuCpSIMR)ScR Complexes 10 Introduction Work in our group with organoscandium complexes has revealed the reactivity of these compounds to be excellent models for Ziegler-Natta catalysis. Our entry into this chemistry began with the permethylscandocene hydrides and alkyls of general formula Cp*2 ScR.1 The use of the bulky Cp* ligand (Cp* = i^-CsMes) has been a general strategy in our group for the steric stabilization of monomeric, electronically unsaturated, early transition metal compounds.2 The 14 electron Cp*2 ScR complexes are monomeric, in contrast to the normal-ring scandocene analogs, which tend to form dimers.2 Permethylscandocene alkyl derivatives were found to be active ethylene polymerization catalysts and were amenable to kinetic studies of chain propagation (olefin insertion) and chain transfer by /9-H elimination.4 As predicted by these experiments and later demonstrated in oligomerization studies, C p*2 ScCH2 CH2 CH 3 was found to initiate true "living" ethylene polymerization at -80 °C (Equation 1). Cp*2ScR n C2H4 (1) Cp*2Sc The insertion of a-olefins by permethylscandocene alkyls does not occur, however. Instead, the scandium activates a vinylic C-H bond of the olefin in a a-bond metathesis reaction to form a scandium alkenyl species and free alkane (Equation 2).1 R* Cp*2ScR Cp*2Sc RH (2) 11 W e attribute this behavior to unfavorable steric interactions between the substituent on the olefin and the (r?5-CsMes) ligand when the olefin * orbital approaches the metal-carbon bond for insertion (A). An orthogonal approach of the olefin to the metal (B) minimizes steric repulsions, favoring the a bond metathesis reaction. A B Efforts to relieve steric crowding at the scandium led to the synthesis of the scandocene complexes {(i;5-CsMe4 )2 SiMe2 }(PMe 3 )ScHl Op(PMe3 )ScH, and [{(»?5C5H3CMe3 )2 SiMe2 }ScH]2 , [DpScHfe. By tying the cydopentadlenyl tings back with a dimethylsilylene bridge, we hoped to increase access of the olefin to the scandium center and in this way promote a-olefin insertion. As expected, these scandium hydride complexes are more reactive than the parent permethylscandocene hydride derivative, catalytically dimerisng a-olefins by the mechanism shown below (Scheme I).® Following the second olefin insertion, 0-H elimination from the tertiary /9-carbon to liberate the olefin dimer and regenerate Sc-H is rapid relative to a sterically disfavored, third a-olefin insertion. 12 Sc H - 12 Sc-H Sc + R H Schema I. P-H elim. R Sc-H + =----- / V etc. Catalytic a-Olefin Dimerlzetlon by (Op)(PMe3)ScH and [DpScH]2 Thus, we were approaching our goal of developing an organoscandium system capable of inserting a-olefins in a multiple fashion. It was clear from our experience with the linked metallocene systems that varying the sterics of the supporting ligand system has a striking effect on the reactivity of the scandium center. This prompted us to examine further variations of the ligand framework. W e anticipated that scandium derivatives of the type {(i79-CsMe4)M e 2 Si(i?1-NCMe3 )}ScR ("(Cp*SINR)ScR") would differ significantly from the scandocene derivatives, since the amido group should render the metal more Lewis-acidic and even more electron deficient, two properties that we felt would encourage olefin insertion. Also, to the extent that the steric bulk of the cyclopentadienyi rings promote chain transfer by yS-H elimination, this Ie3s sterically crowded scandium system would be more likely to effect the multiple insertion of a-olefins. 13 Results and Discussion The Synthesis and Characterization of [(C p*SIN R )(PM e3)S c0i-H )]2 The Cp*SiNR ligand [Cp*SiNR = (r/5-C5Me4)SiMe2 (»r1-NCMe3 )] is prepared straightforwardly by reacting (CsMe^SiMeaCI with lithium 1-butyl amide (Equation 2). + MejCNHU (3) "Cp'SiNR* Double deprotonatlon of the ligand with n-BuU followed by reaction of the dilithium salt of the ligand with ScCl3 (THF) 3 affords (Cp*SiNR)ScCI complexed by variable, nonstoichiometric amounts of THF and UCI. The presence of THF in all of our scandium complexes is undesirable since it blocks the reactivity of the metal by binding tightly to the active site. In this compound, THF is removed by heating the solid (Cp*SiNR)ScCI« (THF)X* (UCI)n at 100*C in vacuo for one day. The lithium halide may be removed by sohxlet extraction of [(Cp*SINR)ScCI]x with hot toluene, although it does not interfere with the subsequent alkylation step. The bulky bis(trimethytsilyl)methyl derivative (3) is the only alkyl derivative that could be successfully prepared from (Cp*SINR)ScCI. The reaction of other alkyllithium reagents such as MeU, PhU and Me3 SiCH2 U with the scandium chloride complex did not afford clean products. Despite several attempts, single crystals of 3 suitable for x-ray analysis could not be obtained and ebulliometric molecular weight analyses were inconclusive in determining its molecularity in solution.6 Compound 3 is hydrogenated cleanly in the presence of PMe3 to form the hydride complex [(Cp*SiNR)Sc(PMea)]2(p-H)2(4) in 20% overall yield starting from ScCl3 (THF)3 (Scheme II). Hydrogenation of 3 in the absence of PMe3 affords a mixture of products. The usefulness of PMea as a soft, weakly binding Lewis base for stabilizing 14 these highly electronically and coordinatively unsaturated organoscandium complexes was discovered in earlier work with {(>;5-CsMe4 )2 SiMe2 }ScR and some mixed-ring complexes, (Cp*)(Cp/)ScR (Cp'= qS-CsHs, f?5 - 1 ,3 ,4 -C 5 H2 Me3 ) . 7 Single crystals of 4 (as toluene-dg solvate) grew in an NMR tube when a concentrated solution of 3 in toluene-dg was hydrogenated. An x-ray structure determination revealed a double hydrogen-bridged dimer of Cz symmetry. As can be seen from the ORTEP drawing in Figure 1.1, the ligand framework of the molecule is reminiscent of that of the bent metallocene systems, although the bite angle of 104.8(3)* is closer to a tetrahedral geometry than the 133.9* ring centroid-Sc-ring centroid angle found in {(^-CgM e^SiM eaHPM ealScH. An extremely long Sc-P bond length of almost 3A is a notable feature of the structure. This distance is half an angstrom larger than the sum of the covalent radii for Sc and P. 8 It is also longer than the Sc-P bond in {(’75-C 5 Me4 )2 SiMe2 }(PMe 3 )ScH (2.75A).® Consistent with the long Sc-P distance, variable temperature 31P NMR studies reveal that PMea dissociation is rapid. The atoms connected to nitrogen are coplanar to within ±0.02A, indicating sp2 hybridization of the nitrogen atom. The Sc-N bond distance of 2.058(3)A is comparable to the 2.049A Sc-N bond length found in tris-(hexamethyidisilytamido)scandium. 9 The SWM bond length of 1.663(3)A is a bit short in comparison to typical values of 1.70-1.78A for related ligand systems, 1 0 indicating that a significant amount of N -S i(p * 1> da) interaction Is competing with nitrogen lone pair donation to the scandium. 1 1 The poor solubility of 4 has prevented determination of its solution molecular weight; however, a monomer/dlmer equilibrium in solution is indicated by the presence of two peaks (ca.4:1 ratio) in the 31P spectrum of 4 measured a t-6 6 *C. The variable temperature 1H NMR study was less informative, revealing only a single, detectable species over the range -80 ”C to 25*C. Figure 1.1. ORTEP drawing of [(Cp*SlNR)(PMe3)ScHl2 (4) with 50% probability ellipsoids. 15 16 NCMe, Lis 25°C, toluene (Cp*SiNR)ScCI»(LiCI)n*(THF)x (1 ) 1. vacuum, 100°C 2. LiCH(SiMe 3) 2 (2) / M e *C H 2f P M e 3 ,PMej SiMe. M e,C Scheme III. (3 ) CMe-i Synthesle of [(Cp*SINR)(PMe3)ScH]2 17 The hydride resonance occurs at S 4.45 in the 1H NMR spectrum and is considerably broadened because of its proximity to the quadrupolar scandium nucleus (I = 7/2,100% abundance, -0.22x102 0 Q/m 2 ) . 1 2 An attempt was made to locate the Sc-H stretch by IR analysis. The Nujol spectrum of 4 includes a broad band of medium intensity at 1152.4 cm ' 1 that is absent in the spectrum of the corresponding deuteride, 4 -d i. If this band corresponds to v Sc-H, the corresponding Sc-0 band should be shifted to ca 817 cm'1. This region of the spectrum is already obscured by other absorption bands, although there does appear to be a broadening and increase in intensity in the group of bands in this region for 4-cfi, perhaps from the presence of an overlapping Sc-0 band. A vibrational energy of 1152.4 cm' 1 is rather low, even for a bridging metal hydride.1® To determine if this could be due to some polymeric composition of the scandium hydride complex resulting from its manner of isolation (precipitation from hexane), a Nujol mull of 4 that had been recrystallized from toluene was examined by IR. No spectral changes were observed. As mentioned, the trimethylphosphine in 4 is quite labile, dissociating reversibly from the metal on the NMR time scale. By contrast, the PMea ligands of Jordan’s [Cp2 Zr(H)(PMe3 )2 ][BPti4 ] 1* do not exchange with free PMea on the NMR time scale. To determine if a bis-phosphine adduct, (Cp*SiNR)(PMe3 )2 ScH, analogous to Jordan's compound might be formed, a tenfold excess of PMea was added to 4 in an NMR tube. No new resonances were detected in this sample by either 1H o r31P NMR from 25*C to -80*C, indicating that coordination of a second phosphine by the scandium does not occur to a significant extent [(Cp*SiNR)Sc(PMe3 )] 2 (p-H) 2 does indeed cleanly catalyze the polymerization of aolefins, albeit rather slowly. For example, with 0.1 mole percent [(Cp*SiNR)Sc(PMea)]2 (/i-H )2 in neat pentene at 25*C, polymerization rates of ca 50 tumovers/Sc/hr are found, and after 19 hrs polypentene with Mn = 3,000 (Xn * 4 3 ) and POI = 2 . 1 (gpc relative to polystyrene) is obtained. Propene, 1-butene, hexadiene and butadiene have also been polymerized with 18 this catalyst system. The polymerization activity appears to be fairly general to monosubstituted olefins (some exceptions will be discussed later in this chapter and in Chapter 2). In order to gain insight into the nature of the active propagating species, we set out to isolate and characterize products resulting from olefin insertion into the Sc-H bond of this compound. These species are discussed in the sections that follow. [(Cp*SiNR)(PM»3)Sc]20i,*j2,r?2-C2H4): An Unusual Ethylene Bridged Dimer from the Stoichiometric Reaction of [(Cp*SiNR)(PMe3 )Sc(/*-H) ] 2 with Ethylene The seemingly simplest place to start in preparing an olefin insertion product from [(Cp*SiNR)(PMe3 )ScH] was with ethylene. Addition of one equivalent of ethylene to 4 afforded an unexpected product, however. Instead of the familiar triplet, quartet pattern of an ethyl group, a singlet at 6 -0.17ppm was observed in the 1H NMR spectrum. When the reaction was performed with 1 3 C 2 H4 , the proton-coupled 13C NMR spectrum of the product revealed a triplet pattern, with a principal Jch coupling of 142Hz. Fine structure from second order coupling effects is also apparent (Figure 1.2). Figure 1A Methylene Region of 1JC NMR Spectrum of [(Cp*S»IR(PMaj))Sclj0i,tiV*1,C2H4) (5). 19 iSc SiM» 2 C 2H4 M0 3 C -C,H, 2r16 MezSI MeaC Scheme III. SiM«» CM*, Reaction Pathway for the Formation of [(Cp*SiNR)(PMe3)Sc]2(n.n2.n2-C2H4) 20 Upon closer inspection of the 1H NMR spectrum, a resonance consistent with free ethane was located at ca 0.75 ppm. The NMR evidence indicated that the reaction shown in Scheme III had occurred to produce the ethylene-bridged scandium dimer [(Cp*SiNR)Sc]2 (At-C2 H4 ) (5). This assignment was somewhat puzzling since the C-H activation of benzene solvent by 4 does not occur readily until the system is heated to 80 °C, whereas this intermolecular activation of a normally less reactive sp3-hybridized C-H bond occurs readily at ambient temperature. To confirm our assignment, an x-ray crystal structure determination was undertaken. As can be seen in the ORTEP drawing in Figure 1.3, an ethylene-bridged scandium dimer is indeed formed. The r)2, t)2 coordination of the ethylene bridge is quite unusual and has been observed in only a few other metal systems. 1 5 The C-C bond length of the ethylene bridge (1.433(12)A) is intermediate between that of a single and a double C-C bond7 and is comparable to the bond lengths observed In transition metal olefin complexes. 1 5 The sp2 hybridization of the ethylene carbons of the C2 H4 bridge is reflected in the H-C-H angle of 120(6)* as well as in the large Jch of 142Hz. As in the structure for the Sc-H dimer, the cyclopentadienyl rings are more or less cis. The Sc-P bond length of 2.825(3)A is significantly shorter than in 4, although phosphine dissociation is still rapid. The amide group is again planar to within ±0 . 0 1 6A. Both the Sc-N bond distance-(2.071 (6 )A) and the Si•> / ' N bond distance (1.722(6)A) are longer than in compound 4. This species has been found to polymerize ethylene as weil as propane. 21 22 Although the formation of 5 was a surprising result, there is precedent for this type of reactivity in homogeneous Ziegler-Natta catalyst systems based on Cp2 TilvX2 and Cp2 Zr*vX2 . 1 7 0-C-H activation to form a dimetalloalkane is believed to be involved in the reduction of Cp2 Ti(IV) to catalytically inactive Cp2 Ti(lll) and is perhaps responsible for the aging process of titanium based heterogeneous Ziegler-Natta catalyst systems as well. Since Zr(IV) is more resistant to reduction than Ti(IV), ethylene-bridged products analogous to S could be isolated from Cp2 ZrX2 /Al2 Ets mixtures and were structurally characterized (Figure 1.4).1® / Cl Cp2ZrCl2 Et + AI2Ete Et Figure 1A. Ethylene-Bridged Dimers isolated from CpjZrClj / AI^Me, Mixture 23 To elucidate the mechanism of formation of 5, the reaction was followed by low* temperature 1H NMR spectroscopy. We found that at ca 0 °C all the ethylene (one equivalent per scandium) was consumed to form a new species, presumably the ethylene insertion product. Because of the fluxionality of this intermediate, a clear spectrum was obtained only by cooling the NMR sample to *90 aC. In addition to the signals for the Cp*SiNR ligand and PMe3 , a broad multiplet at ca 6 0.0 corresponding to the a-methylene and a multiplet at ca S 0.96 for the £-methyi group were observed. When the sample was allowed to warm to 25 *C, the insertion intermediate converted to the ethylene-bridged product, 5. The production of half an equivalent of ethane per equivalent of ethylene in this reaction was confirmed by a Toepler experiment, in which 95.5% percent of the theoretical amount of ethane was collected. The 1H NMR spectrum of compound 5 displays interesting variable temperature behavior. As in the hydride dimer, PMea dissociation is sufficiently rapid that the diastereotopic methyl substituents on the cyclopentadienyl ring and on the silylene bridge of 4 are equivalent in the room temperature 1H NMR spectrum. A slower process resulting in an averaging of the signals of the inequivalent methylene protons of the -CH2 CH 2 - bridge is also observed (Figure 1.5). * J * b) i i o j "~"i i 1 i Figure 1.5. a) 400 MHz 1H NMR spectrum of ethylene bridge hydrogens in compound 5 (frozen out* at ca. -90°C, b) Spectrum after coalescence, (ca. -40°C). Asterisked peak is an impurity. 24 The exchange of environments of these protons requires the ethylene bridge to, in effect, undergo a 180* flip about the carbon-carbon axis. The mechanism by which this rearrangement occurs is currently unknown. A diastereotoplc site exchange rate constant (^exchange) at 274 *K of 1.6x103 sec' 1 was determined by using coalescence methods. This value is unchanged by the addition of excess phosphine to the solution of the complex. The rate constant corresponds to a AG* of 1 2 kcal/mole for the exchange at this temperature. The Isolation and Structural Characterization of PMoa-free Scandium Alkyl Dimers, [(Cp*SINR)Sc]2(M-alkyl)2 When 4 is reacted with an equivalent of propene per scandium, a PMea-free scandium propyl derivative (6 ) is formed (Equation 4). An x-ray structure determination of 6 revealed it to be a dimer in the solid state with two three-center, two electron alkyl bridges. A similar PMe3 -free scandium n-butyl derivative (7) is obtained when 1 -butane is used instead of propene. Loss of PMea <n this case is less facile, so that a few cycles of redissolving the product in toluene and then stripping off the volatlles is necessary to completely remove the PMe3 . A double alkyl-bridged dimer is presumably formed in this case as well. MsjSI SIMa, PMa, / M«jC ^ CMa, 4 -2PMaj 2 aquiv propan#; -80°C 6 28 As can be seen from the ORTEP drawing of 6 (Figure 1.6), and in contrast to 4 and 5 , the two halves of the dimer are related by a center of inversion. The Sc-C-Sc bridge angle of 89.4(2) ° is considerably more acute than the corresponding Sc-H-Sc bridge angle in 4 (114.1 °). This reflects the higher directionality of the carbon sp3 orbital as compared with the s orbital of the hydride, such that a narrowing of the bridge angle is necessary to maximize orbital overlap. Similar structural trends are evident in other p-H and /j-alkyl systems. For example, the 114(3)° Y-H-Y* bridge angle in [(CH3 C 5 H4 )2 YH(THF)]2 1 2 contracts to 87.7(3)° in [Cp2 YMe]2 , 1 9 angles surprisingly similar to those found in the scandium systems described here, considering yttrium’s larger size. Likewise, the n-H bridge angle of 103(2)° in the dimethylaluminum hydride dimer2 0 contracts to 74.7(4)° when the hydrides are replaced by methyl groups in the trimethylaluminum dimer.2 1 As expected, the atoms connected to the amide nitrogen are coplanar to within ±0.01 A. The bite angle of the chelating Cp*SiNR ligand (102.5°) is slightly smaller than in the Sc-H dimer, and the Sc-N bond distance (2.083(5)A) is even longer than in the ethylenebridged complex 5, although the Si-N bond distance (1.722(6)A) is about the same. 1H and 13C NMR data for 6 are consistent with its solid state structure. In the 1H NMR spectrum, a broad pseudotriplet (3 Jhh - 3*6 Hz) is observed upfield at -0.11 ppm for the methylene group next to scandium along with a triplet at 6 1.06 (3 Jch = 6 . 6 Hz) for the end methyl group. The /9-methylene protons form a complex pattern that is partly obscured by the the 1-butyl amide resonance. This complexity may be attributed to the diastereotopic nature of the two hydrogens as well as to their coupling to both the a-methylene and 7 * methyl hydrogens of the propyl ligand. An almost identical 1H NMR spectrum is observed for the n-butyl derivative, with the exception of an extra multiplet at 1.38 ppm for the 7 methylene hydrogens. The 1H NMR spectrum of the propyl derivative does not change from 25°C to -80°C in toluene-cfe- Upon addition of an equivalent of PMeg to the NMR tube, however, an entirely 27 new set of resonances is observed, probably as a result of PMe3 coordination to form (Cp*SiNR)(PMe3 )ScCH 2 CH2 CH 3 . Site exchange of the PMe3 in the propyl derivative is rapid, so that at -80 °C the methyl resonances of the Cp*SiNR ligand are broadened but not resolved into their diastereotopic components. Interestingly, excess PMe3 appears to enhance the rate of PMe3 dissociation, such that the ligand resonances are sharp (i.e., in the fast exchange limit) even at -80 sC in the presence of five equivalents of PMes. As is common to all our scandium alkyl complexes, the 13C resonance for the acarbon in both the n-propyl and the n-butyl derivatives is broad and shifted considerably downfield (6 50.52). This signal appears as a triplet in the proton-coupled spectrum. The five-coordinate bridging geometry of the a-carbon is reflected in the lo w 1 Jch of 103Hz since in non-bridging Sc-alkyl systems sp3 couplings of ca 118-125 Hz are typically observed.2 2 There is no evidence for an agostic C-H interaction in these complexes in either the crystal structure of 6 or the IR spectra of 6 and 7. Unfortunately, the low solubilities of both the Sc-n-propyl complex and the Sc-n-butyl complex in benzene prevent their solution molecular weight determination. As discussed in Chapter 2, the absence of PMea makes these systems more active olefin polymerization catalysts than the Sc-H dimer. Reaction of [(Cp*SINR)(PMo3 )ScG*-H) ] 2 with the Qem-Dlsubstltuted Olefin 2-MethyMPentene While dimerizatlon serves as a driving force for PMe3 loss in the Sc-n-alkyl derivatives mentioned above, bulkier derivatives such as the 2 -methylpentyl derivative do not form /j-alkyl bridges as readily and therefore retain PMe3 in their coordination sphere. (Cp*SiNR)(PMe3 )ScCH2 CH(CH 3 )CH 2 CH 2 CH 3 (8 ) was prepared by treating 4 with an excess of 2-methyM-pentene (Equation 5). Our interest in this derivative arose from the question of whether branching can occur in a-olefln polymerization by reinsertion of the olefinic polymer 28 chain ends (formed in chain transfer by f)M elimination) into the propagating metal alkyl. Since 8 does not react further with 2 -methyl-1 -pentene, the formation of branched polyolefin in the polymerizations is unlikely. / (Cp*StNR)(PMes)ScH + l = / (Cp*SfNR)(PM*j)i CHjCHjCHj Although a solution of 8 yields a gold-brown oil upon removal of solvent, cooling a concentrated solution of the compound in pentane affords white microcrystalline material in reasonable yield (50%). Because of the fluctionality of 8 , presumably involving phosphine site exchange, its room-temperature 1H NMR spectrum is quite broad and uninterpretable. Upon cooling the sample to -80 *C, the spectrum becomes clearer, revealing the presence of two species in roughly equal amounts. It is likely that the two sets of resonances belong to two diastereomers since there are two stereocenters in the system (the Sc center and the pcarbon of the alkyl substituent). The signals of the alkyl substituent are difficult to sort out; however, hydrogenation of 8 affords 1.1 equivalents of 2-methylpentane relative to Sc, as measured by 1H NMR (using hexamethyidisiloxane as a standard). The formulation of 8 is also confirmed by elemental analysis. Ebulliometric molecular weight analysis of the species in solution was unsuccessful, presumably because of the dissociable PMe3 , and after standing for a week in benzene solution, the complex had decomposed. As expected, although 8 does not further insert 2 -methyl-l-pentene, it does polymerize propene. 29 The Reaction of Styrene with [(Cp*SiNR)Sc(PMe 3 )] 2 (^-H ) 2 When an excess of styrene is added to the scandium hydride complex 3, instead of the formation of polymer, the reaction stops after two insertions. The product is isolated as a bright orange-yellow solid (9) in 80% yield. Cryogenic MW analysis of the product indicated that it was a monomer and that two styrenes had been incorporated by 4 (Exptl MW: 596 ±36; Calcd MW: 580). Efforts to grow crystals of this material for an x-ray structure were unsuccessful; however, the coupled 13C NMR spectrum of material is consistent with the structure shown in Equation 6 . Ph [(Cp*SiNR)(PMe3 )ScH]2 excess styrene -------------------------------(Cp*SiNR)(PMe3 ) S c ^ / V ^Ph Deuterium-labeling studies were also performed but were difficult to interpret. The orientation of the second inserted monomer was probed by cleaving 9 with D2 and locating the deuterium in the organic product by 2H NMR. Likewise, the orientation of the first inserted monomer was probed by examining the 2H NMR of the double-insertion product starting from the scandium deuteride, 9-dv In both cases the major deuterium resonances ' i appeared to be benzyllc, appearing at ca 5 2.4 in the first case and ca 6 2.6 in the second; there were also broad resonances upfield of these signals, which might be assigned to nonbenzyile positions. Thus, while the primary scandium product contains head-to-head coupled styrenes, some minor regioisomers may be formed as well. This is also indicated by the 13C NMR of a second crop of isolated product, which contains a few small peaks besides those of the major isomer 9. The head-to-head coupling of the two styrene molecules contrasts with the highly regioregular head-to-tail coupling observed with propene and may be attributed to the ability of the phenyl substitutuent to stabilize the partial (6 ) 30 negative charge at the incipient a-carbon in the transition state for olefin insertion.4 , 2 2 , 2 4 Steric interference from the "2-1" orientation of the second inserted styrene unit may be the cause for reaction stopping after two insertions. Reaction of 4 with a single equivalent of styrene was not clean and appeared to produce a mixture that contained some of the double insertion product 9. Synthesis of the Chiral System (tBuCpSiNR)ScR Among our goals in studying these Ziegler-Natta model systems is to understand the detailed interactions responsible for stereospecific olefin insertion to form either isotactic or syndiotactic polymer. We have therefore embarked on further modifications of the basic silylene-bridged, monocyclopentadienyl-amido ligand-framework in order to study the stereodirecting effect of the ligand on the polymerization. As demonstrated with Brintzinger's ansa-metallocene complexes,2 ® homogeneous Ziegler-Natta catalysts with chiral ligand frameworks can be effective at producing highly isotactic polymers. There are a number of ways to modify the Cp*SiNR ligand to make the scandium complex chiral. Asymmetry may be introduced at the amide group, at the silylene bridge or at the cyclopentadienyl ring. We thought that an asymmetrically substituted cyclopentadienyi ring would have the greatest influence on the active site of the metal. Our first effort in this direction has been the synthesis of the chiral system (tBuCpSiNR)ScR [,BuCpSiNR = {(»;5 -C 5 H4 CMe3 )Me2 Si(»71 -NCMe3 )}]. Preliminary results of our synthetic efforts with this new ligand system are discussed below. The ligand (CsH4 CMe3 )SiMe2 (NCMe3 ) is prepared in the same manner as its tetramethylcyclopentadienyl analog. (CsH4 CMe3 )SiMe2 CI is prepared by a single ring substitution on SiMe2 Cl2 and the remaining chloride is metathesized by U(NHCMe 3 ) (Equation 7). Deprotonatlon of the ligand with two equivalents of n-butyllithium in petroleum ether affords the dimetallated ligand, not as a powder, but as a polymeric material, which is isolated as a glassy solid upon complete removal of solvent. Although the 1H NMR spectrum of U 2 (tBuCpSiNR) in benzene-cfo is quite broad, a clear spectrum is produced upon reaction of the ligand with ScCl3 (THF) 3 to form (tBuCpSiNR)(THF)ScCI (12). Removal of THF from this compound is not accomplished as easily as for (Cp*SINR)ScCI« (THF)x* (UCI)n- Heating the compound in vacuo to temperatures above 180*C is necessary before THF is liberated and complete removal of THF by this method is difficult. Replacement of the chloride with bulky ligands (e.g., -CH(TMS)2 ) and chelating ligands (e.g., acetylacetonate) does not displace the coordinated THF. Reaction of 12 with UCH 2 PMe2 in toluene, however, affords a precipitate that analyzes reasonably well for THF-free [(Cp*SINR)ScCH2 PMe2 ]x (13) coprecipitated with UCI. The 1H NMR spectrum of 13 in THF-ds Is complicated and indicates the presence of a mixture of species. Although 13 cannot be hydrogenolyzed, it does react with an equivalent of HCI gas to afford (tBuCpSINR)(PMe3 )ScCI (14). The chloride derivative 14 is isolated in 53% overall yield based on ScCl3 (THF) 3 starting material (Scheme V). Ebulliometric analysis indicates that 14 is a monomer (Calcd MW: 406.1; Exptl MW: 476). Heating 14 to 120aC in vacuo for a few hours liberates PMe3 to afford [(tBuCpSINR)ScCI]x (15). A series of NMR tube reactions revealed that 14 reacts with a variety of alkyllithium reagents to cleanly form the phophine-coordinated scandium alkyl derivatives. We were able to isolate crystals of (^CpSINRHPMeslScCHzSiMea (16) for an x-ray structure analysis. 32 / SL NCM63 S c CI3(TH F)3 + 25°C, toluene (,BuC pSiN R )(THF)ScCI (1 2 ) U CH 2 PM 0 2 ; -TH F [(^ C p S iN ^ S c C H zP M e jJ , + UCI (1 3 ) HCI (14) Scheme V. Synthetic of (tBuCpSINR)(PMe3)ScCI 33 (,BuCpSiNR)(PM*])ScCI UCH^SiM*]) MajSIl .Sc -.~^,CH2TMS N (8) VPM . *, The ORTEP drawing of the complex is shown in Figure 1.7. The complex is a monomer with a distorted, three-legged piano stool structure. Only one of the two possible diastereomers, with the PMe3 ligand positioned away from the t-butyl substituent on the Cp ring, is present in the structure. The bite angle (104.1 *) of the ^ C p S iN R ligand is comparable to that of the Cp*SiNR system. The Sc-P bond length of 2.767(1) is shorter than that found in the two structurally characterized phosphlne-coordinated (Cp*SINR)Sc complexes (4 and 5). Once again, the amide nitrogen is planar, and the Sc-N (2.044(3)A) and Si-N (1.731 (3)A) bond lengths are comparable to those found in (Cp*SiNR)Sc complexes. Preliminary observations indicate that the PMea in 16 is non-labile and that the single dlastereomer of 16 found in the crystal structure is formed directly from a single dlastereomer of the chloride derivative 14. For instance, reaction of 14 with phenyllithium produces only a single diastereomer of (tBuCpSiNR)(PMe3 )ScCeHs (*H NMR). When a benzene-dfe solution of this "kinetic1* isomer is heated to 80*C , it forms ca a 13:1 mixture of diasteriomers, in which the second, thermodynamically preferred isomer predominates. Heating a solution of either 14 or 16 in CeDe at 80*C does not lead to any new resonances that might be attributed to the second diastereomer. A strong thermodynamic preference for the one isomer is likely in these cases. Addition of PMe3 to the phosphine-free chloride derivative, 15, regenerates the 1H spectrum for 14; again only the single diastereomer is observed. Figure 1.7. ORTEP drawing of (Cp*SlNR)(PMe3)ScCH2TMS (16) with 30% probability ellipsoids. 34 35 The inability of compound 16 to polymerize propene is attributed to the tight binding of PMe3 by the complex. Polyethylene is formed with this system; however, only a small percentage of the material initiates the polymerization. When either (tBuCpSiNR)(PMe3)ScCI (14) or [(,BuCpSiNR)ScCI]x (15) is reacted with MeLi, a material analyzing correctly for [(,BuCpSiNR)ScMe]x (17) is formed (Equation 9). The tendency of this material to precipitate from a benzene solution has prevented the measurement of Its solution molecular weight The compound exhibits a broad, ill-defined 1H NMR spectrum; however, upon addition of PMea to the solution, a sharp 1H spectrum consistent with the species (tBuCpSiNR)(PMe3 )ScMe is observed. Apparently, [(,0 uCpSiNR)ScMe]x is an oligomer that can be dissociated by the coordination of PMe3. While [(tBuCpSiNR)ScMe]x polymerizes a-olefins (albeit slowly), addition of an equivalent of PMe3 to the complex shuts down polymerization completely. tau 120°c - vacuum ( CpSiNP.)(PMe3)ScCI ------ — ------------PM 0a MeLi -PMe: (^CpSiNRJScCI (9) MeLi [(®uCpSIN R)ScMe]x Most of the phosphine-free alkyl derivatives of this scandium system appear to form oligomers, since intractable oils with complicated 1H NMR spectra are generally obtained. As with [(tBuCpSiNR)ScMe]Xl addition of PMe3 to these products results in a clear 1H NMR spectra. Thus, while a number of phosphine-free alkyl derivatives of the type [(tBuCpSiNR)ScR]x may be prepared, they are ill-defined; hence, their utility as model systems is limited. Although the synthetic possibilities using this ligand system have not 36 a bis(t-butyl)cyclopentadienyl amido complex, would be a more suitable target. Summary Synthetic entry into the monocyclopentadienyl amido complexes (Cp*SiNR)ScR and (tBuCpSiNR)ScR has been described. Because of the reduced steric bulk of the ligand framework in these compounds, the scandium is more open to external substrates. The increased "exposure" of the scandium center serves to enhance its reactivity, hence the ability of these complexes to polymerize a-olefins. It also necessitates the development of special synthetic routes and “back-door" approaches for removing tightly coordinating Lewis bases that satisfy the demand of the coordinatively and electronically unsaturated scandium for electrons. Thus, there is a tradeoff between reactivity and ease of synthesis in working with these highly Lewis acidic organoscandium complexes. Historically we have used the bulky Cp* ligand in our labs to sterically stabilize highly unsaturated transition metal centers against the formation of unreactive oligomers. We have found, however, that despite the tendency of the less bulky monocyclopentadienyl amido scandium complexes to oligomerize, these complexes display greater olefin-insertion activity than our original monomeric Cp^ScR complexes Therefore, our present strategy with scandium is to fine tune the sterics of the ligand environment to allow as much access of substrate to the metal as possible while preventing the formation of unreactive polynuclear compounds. 37 Table I. 1H NMR Data for (Cp*SiNR)Sc Complaxaa.* (CH3)4CpH (CH3 )4 CpH (CH3)3CN (CW3)2Si 2 .0 [(Cp*SiNR)ScCI]x (2) (90 MHz, CeDe) [C5 (CH3 ) 4 Si] (CH3)3CN (CH3)2SI 2.0 s, 1.9 s 1.3 s 0.7 s (Cp*SiNR)ScCH{SI(CH 3 )3 > 2 (3) (400MHz, C6 D8) [C5 (CH3 ) 4 SI] (CH3)3CN (CH3)2Si CH{Si(CH 3 ) 3 > 2 CH{SI(CH 3 ) 3 } 2 2.05 s, 1.91s 1.3 s 0 .6 6 s 0.14 s -0.57 s [(Cp*SiNR)(PMe3 )Sc]2 (p-H) 2 (4) (400 MHz, toluene-cfa) [C5 (CW3 )4 Si] (CH3)3CN (CHaJzSi (CH3)3P Sc-H 2.21 s, 2.06 S 1.28 s 0 .6 8 s 0.89 s 4.45 s. br [ (Cp*SiNR) (PMe3 )Sc]2 (p*H) 2 (4) (400 MHz, toluenes/g, -80 *C) [C5 (CH3 ) 4 Si] (CH3)3CN (CH3 )2 SI (Ch 3)3p S&H 2.36 s, 2.16 s 1.99 s, 1.90 s 1.31 S 0.84 s, 0.74 s 2JpH = 4 0.89 d 4.15 s, br [(Cp*SiNR)(PM«3 )Sc]2 (M-C2 H 4 ) 2 (5) (400 MHz, toluene-cfe) [C9 (CH3 ) 4 SI] (CH3)3CN (CH3 )2 SI <c h 3)3p C2 « 4 2.33 3,1.89 s 1.24 s 0.79 s 2JPH = 3.9 0 .8 8 d -0.17 s [(Cp*SiNR)(PMe3 )Sc]2 (M-C2 H 4 ) 2 (5) (400 MHz, toluenes/e, -90 *C) [Ce(CH3 )4 SI] 2.46 S, 2.35 3 , 1.89 8,1.86 s 1.3S 0.98 s, 0.90 S 2J rh = 3.9 0 .8 8 d 0.18 d, br, -0.5 (Me4 CpH)SiMe2 (NCMe3 ) (90 MHz, C6 D6) (CW3)3CN (C« 3 )2 SI (CH3)3p C2H4 s, 1 . 8 s 2.7 1 .1 0 .1 s s 38 [(Cp*SiNR)Sc]2 (M*CH2CH2CH 3 ) 2 (6) (400 MHz, CeDe) [C5(CH3)4 Si] (CH3)3CN (CH3)2Si CH2CH2CH3 CH2CH2CH3 CH2CH2CH3 2.17 s, 1.91 s 1.25 s 0.72 s -0.111 3JHh = 8.6 ca. 1.5mb, ca. 1.2 mb 1.061 3Jhh = 6.6 [(Cp*SiNR)(PMe3)ScCH2CH2CH3 (400 MHz, toluene-cfe) [C5(CH3)4Si] (CH3)3CN (CH3)2SI P(CH3)3 CH2CH2CH3 CH2CH2CH3 CH2CH2CH3 2.10 s, 2.05 s 1.4 s 0.66 s 0.74 br -0.094 m ca. 1.67 m 1.15- t [(Cp*SiNR)Sc]2(/i'CH2CH2CH2CH3)2 (7) (400 MHz, CfiOg) [Cs(CH3)4SI] (CH3)3CN (CW3)2Si 0.74 s CH2CH2CH2CH3 CH 2 CH2 CH 2 CH 3 CH2CH2CH2CH3 CH2CH2CH2CH3 2.18 s, 1.92 s 1.28 S [C5(CW3)4Si] (a: 2.43 s, 1.90 3) (b: 2.33 s, 2.34 s, 1.85,1.84) 1.47 (a: 0.88 s) (b: 0.704 s, 0.698 s) 0.5 d 2J ph =4.4 a&b: (0.14 m, -0.17 m; -0.08 m) (Cp^SINRJfPMesJScR (8)d R = CH2CH(CH3)CH2CH2CH3 (500MHz, toluene-cfe. -80 aC) (CW3)3CN (CH3)2Si (CW3)3P S cC H r (Cp*SiNR)(PM* 3 )ScR’ (9) R’ = CHfCgHslttfeCH^htetCsHs) (400MHz, CeOg) [C5(CH3)4SI] (CH3)3CN (CH3)2SI (fi,y ,6 )C H 2 a-CH2 aryl region -0.111 3JHH - 9 ca. 1.4 mb, ca. 1.2 mb 1.38 mc 0.961 3Jhh = 7.2 2.25 s, 2.10 s, 2.06 s, 1.93 s 1.27 s 0.77 s. 0.64 s (1.85 m, 1.65 m) 2.69 m, 2.02 m* not assigned 6.53 d, 5.591, 7 .3 1, 7.0-7.2 muttlplets a1H (90Hz, 400 MHz and 500MHz)NMR spectra were obtained at ambient temperature unless otherwise stated. Chemical shifts (£) are reported in ppm using either tetramethylsilane or residual protons in the solvent as an internal reference. bOverlap with resonance for (CH3)3CN precludes accurate assignment of these signals. 39 cSignal appears as a quartet overlapping the fi-CHz resonance. dTwo diastereomers (a and b) are present in roughly equal amounts. The chemical shifts of the alkyl substituent could not be assigned with confidence. eAbsolute assignments have been based on chemical shift and coupling behavior and have not been confirmed by a homonuclear decoupling experiment. 40 CO CM sa . Complexes.1 3 O O tor (Cp'SINR)Sc ■*r CM O x ’ IT x CM •O’ CM to ’IT II x o o o co CM to CM T— CM X X o JCH =118 X O ' co co co a U* §,K> T- S OJ CO 5? ■o- co QO IO c\i d r* CO o to o to co co c to § .21 S' jo, .2 1 ? % X % X S « </) s 0 1 ? 2^ E f ? 2Og s C CLJO 2 O 9 a 1? 2 2 co g a. .2 O o CM CM 00* CM t“ rT- 35 Q 2 i 2 is 3 rt ?i n o CM S-r .S> O o CO CO to to ill O 9 1? Q 8.0 to O co (CH3)2Si Table II. 13C and 31P data CM C5(CH3)5Si C 5(CH3)5Si 14.9; 14.6; 12.6; 11.8 108.1 J ch =125; 125; 125 41 42 CO N I fQ. 3 O O CO CM T~ II S2 n 81 ii X X X o o o —} CO CO CO a So in Co II X o II II X X o o o “> -a -9 CO in II X o 03 CO CO E & CO c .S> < ■A in CO O) tn in CO CM ■O' % in in j— co II CL £ £ CM CO CO r- co r“ in r* CM y— r>. in CM *— CO CM y— CO o CO 43 Table III. 1H and 31P NMR Data for (tBuCpSiNR)Sc Com plexes.* Compound (lBu-CpH)SiMe2CI (1H: C8D6, 400MHz) QBu-CpH)SiMe2 NCMe3 (1H: CgDo. 500MHz) Assignment QBu-C5H3)H ttBu-C5H3)H (CW3)3C (CH3)2SI QBu-C5H3)H QBu-C5H3)H (CH3)3C (CH3)3CN (CH3)2SI 6.62 d, 6.4 d, 6.05 s 3.6 s 1.12s 0.06 s, 0.07 s 6.5 d. 6. d, 6.0 s 3.1 s 1.1 s 0.9 s -0.2 s (,BuCpSiNR)(THF)ScCI (11) (1H: C6Dfl, 500MHz) [C5H3(C(CH3)3)Si] 6.42 m, 6.36 m, 6.20 m [C5H3(C(CH3)3)SI] 1.46 s (CH3)3CN 1.32 s (CH3)2SI 0.69 s, 0.65 s C4W80(THF) 3.46 m, 1.03 m (,BuCpSiNR)(PMe3 )ScCI (13) (1H: CfiDg. 500MHz) [C5H3(C(CH3)3)SI] [CsH3(C(CH3)3)Si] (CH3)3CN (CH^zSI (CH3)3p 6.52 m. 6.37 m, 6.03 m 1.44 s 1.32 s 0.54b s 0.68 d 2J ph = 5.7 (31P: CqDs , 36.2MHz) (CH3)aP -44 [(,BuCpSiNR)ScCI]x (14) ( 'H: CsDe. 90MHz) [CsHW CtCH^SIJ 6.67 m, 6.60 m, 6.22 m 1.35 1.31 0.70,0.57 (,BuCpSiNR)(PM*3 )ScCH2TMS (15) (1H: CeDs. 400MHz) [CsHsW CHaW Si] [CsH-jW C H jM S I] [C5H3(C(CH3)3)SI] (CH3)3CN (CH3)2SI (CW3)3P Crt2SiMe3 6.47 m, 6.26 m, 5.8 m 1.46 s 1.40 s 0.72 s, 0.55 s 0.59 d 2J ph = 5.4 0.16 d, 0.60 d JCH = 11 (CH3)3P -48 (CH3)3CN (C W s^l (31P: toluene-dg, 36.2MHz) 44 [(tBuCpSiNR)ScCH3]x (16) (*H: CeDe, 400MHz) [C 5 H3(C(CH3)3)Si] [C5H3(C(CH3)3)Si] & (CH3)3CN (CH3)2Si ScCW3 6.4 br 1.42 br 0.75 br -0.27 (,BuCpSiNR)(PMe3)ScCH3 (1H: CeDe. 400MHz) [C5H3(C(CH3)3)Si] [C5H3(C(CH3)3)Si] (CH3)3CN (CH3)2SI CH2SiMe3 (CH3)3P ScCW3 6.42 m, 6.29 m, 5.74 m 1.36 s 1.43 s 0.68 s, 0.49 s 0.16 d, 0.60 d JcH = 11 0.71 d 2J ph = 3.8 -0.45 (31P: totuene-de. 38.2MHz) (CH3)3P -51 ^ “CpSiNRXPM ^ScCeHs (17a)° (1H: CeDe, 400MHz) [C5H3(C(CH3)3)Si] [C5H3(C(CH3)3)Si] (CH3)3CN (CH3)2Si (CH3)3P ScCfiHs 6.6 m, 6.5 m, 5.9 m 1.27 s 1.15s 0.81 s, 0.59 s 0.68 d 2J ph = 5.2 7.49 d, 7.291, 7.22 »t (tBuCpSINR)(PMe3)ScCeH5 (17b)° (1H: CeDe. 400MHz) [C5H3(C(CH3)3)SI] [CsH3(C(CH3)3)Si] (CH3)3CN (CH3)2Si (CH3)3p ScCsH9 6.6 m, 6.3 m, 5.9 m 1.46 s 1.29 s O ^ s 0.68 d 2J ph = 5.2 7.49 d, 7 .291, 7.22 =t a1H and 31P NMR spectra were obtained at ambient temperature unless otherwise stated. 1H chemical shifts (S) are reported in ppm using either tetramethylsilane or residual protons in the solvent as an internal reference. 31P spectra are referenced to an H 3 P0 4 external standard. bOther resonance for (CH3)2Si believed to be obscured by P(CH3)3. cChemlcal shift data for both diastereomers are reported. Isomer a is the kinetically formed isomer and is isolated from the reaction of 13 with PhLi. Isomer b is the thermodynamically preferred isomer (ca 13:1), which has not been isolated but is formed upon heating a in solution. 45 Table IV. Crystal and Intensity Collection Data for [(Cp*SINR)(PMe3)ScH]2 (4). Formula: ScPSiNCijH3rC7Hg Formula Weight: 463.66 Crystal Color pale yellow Habit: irregular a=21.238(3)A 6 = 11.470(2)A 8 s 113.16(1)* c = 22.253(3) A v = 4984.0(14) A3 z= 8(4 dimers) density = 1.24 g cm-* X=0.71073 A T: 22* Graphite monochromator Space group: C2/c Absences: hkl, h + k = 2 n + l; h 0 1 ,l= 2 n + l Crystal Sim: .11X.13X.23 mm U»4.26cm‘i (urm*x=0.122) CA IM Diffractometer uscan 20 range: 2*-40* Octants collected: ± h , ± k , 1 Number reflections measured: 8166 Number of independent reflections: 2313 Number with F^*>0: 2164 Number with Fo>>3o(Fo2): 1678 Goodness of fit for merging data: 0.97 Final R-indsx: 0.051 Final goodness of fit: 1.53 46 Tablt V. Final Parametara for Compound 4 x, y, x and U,n* x 104 A to m 2 y x Utq ot B Se 847(.3) 5O0(.6) 2850(.3) 387(2) si 2032(1) -263(1) 2662(1) 504(3) p 1073(.5) 258(.0) 4263(.5) 556(3) N 1388(1) -858(2) 2803(1) 443(8) Cl 1782(2) 1286(3) 2676(2) 438(11) C2 1831(2) 1883(3) 3278(2) 456(12) C3 1346(2) 2758(3) 3104(2) 466(12) C4 080(2) 2703(3) 2305(2) 470(13) C5 1251(2) 1800(3) 2132(2) 463(11) Cl 2370(3) 1772(7) 3072(3) 714(10) C8 1330(4) 3688(6) 3500(4) 812(22) CO 477(3) 3558(6) 1073(4) 706(20) CIO 1021(4) 1547(7) 1383(3) 767(20) C ll 2S74(3) -548(7) 3314(5) 1107(32) C12 2006(5) -685(7) 1844(5) 1055(24) CIS 1827(3) -418(0) 4870(4) 024(22) C14 464(3) -656(6) 4404(3) 722(17) CIS 1011(4) 1562(5) 4724(3) 703(10) CIS 1282(2) -2130(4) 2815(3) 618(13) C17 835(4) -2364(5) 3186(4) 855(22) CIS 050(4) -2604(6) 2088(4) 036(23) CIO 1023(3) -2701(6) 3201(5) 1127(28) 47 T a b le V . (Cont’d) C 20 5041(10) 1007(0) 350(6) 1049(48) C21 5555(6) 585(12) 276(7) 1664(45) C 22 5520(7) -3 7 0 (1 4 ) -0 8 (6 ) 1433(40) C23 5007(0) -8 4 3 (1 4 ) -2 2 5 (8 ) 1711(63) SSe —28(12) 588(20) 2971(12) 3.3(6) E , in , (•? •;« * •*,)) *I«otropic displacem ent p i n i M i t f , B 48 Table VI. Hydrogen Parameter* for Compound 4. x, y and x x 104 Atom X y X B H7A H7B H7C H8A H8B H8C H9A HOB HOC H10A H10B H10C H11A HUB H llC H12A H12B H12C H13A H13B H1SC H14A HUB H14C H15A HUB HlftC H17A H17B H17C H18A H18B H18C H10A H10B H10C H20 H21 H22 H2SA H2SB H23C 2745(21] 2484 21 2282 20 1858 21 1134 29 1104 18 434 21 83 26 557 22 835 27 932 27 1313 10 3118 21 3001 25 2884 25 1865 25 2282 24 2205 22 2159 26 1804 29 1818 20 576 20 501 24 35 21 1363 21 1077 19 •11 24 1066 23 707 10 461 10 560426 873 21 1224 21 2138 u 1820 16 2101 » ! 5060 SOU 5926 5827 •240 6327 2088(34] 1070(34 2110(34 4254(42 3501 46 4148 33 4158 39 3302 42 3854 40 1862 47 777 45 1686 33 -0 35 -1220 46 -581 47 -770 48 -1333 41 -147 37 136 30 -1021 47 -628 37 -880 34 -1380 38 -346 36 2066438 1320436 1063 42 -2301 36 -3187 36 -1926 36 -21804 44 -3304 37 -25204 32 -2806 30 -3582 35 i -2404 48 1712 067 -596 -1200 -1306 -245 3072(20 4125(21 4383 20 3700 21 3876 27 3309 18 2263 21 1733 25 1604 22 1120 26 1293 26 1216 10 3346 23 3181 26 3813 23 1440 26 1007 26 1686 23 4033 23 4600 26 5304 21 4866 21 4305 23 4007 20 4774 22 5206 21 4477 23 3668 21 3148 20 2000 10 1774 27 2001 22 1862 22 2668 23 3 2 a 10 3733 20 586 488 -172 -633 114 -217 6.8(13 6.8(17 8.3 14 0.2 15 13.0 25 1.9 13 0.1 17 10.1 21 9.8 18 11.8 24 11.5 23 6.3 13 6.0 16 12.0 21 11.4 23 12.0 26 0.1 10 8.0 16 0.8 18 11.0 20 9.0 15 7.0 14 8.5 19 8.3| 15 8.3 16 8.6 14 10.3 19 7.8 16 7.8 13 6.0 14 12.2 24 8.0 16 6.4 15 0.2 18 6.7 13 14.4 28 18.3 15.8 13.6 16.7 16.7 16.7 49 Table VII. Anisotropic Displacement Parameters for Compound 4. A tom Un Un Uu Sc Si P N Cl C2 C3 C4 C5 C7 C8 C9 C IO 310 (4 ) 430 7) 479 7) 340 17) 303 23 311 24 400 (27 341 27 412 26 501 35 704 45 560 40 610 42 411 33 00 37 667 38 005 45 30 52 61 50 100 '•7] 100 21 L) 406 4 550 8 687 O 384 21 508 26 476 20 378 26 400 28 523 28 807 51 483 43 802 46 1131 58 8341 54 712 50 1327 68 850 48 848 44 450 30 551 43 677 48 603 43 1161 86 737 6 6 1306 96 1473 1411) 462(5) 877(10) 400 >7> 652 23) 559 30 558 34 600 38 651 38 543 34 706 46 1283 64 1010 53 646 46 2150 101t) 1663 SO] 757 47 741 47 660 45 0 0 6 43 12061 ra 1335 60 2066 101I) 1757 83 1707 8 1 1155 65 1715 146) C ll C12 C IS C14 C IS C IS C17 C IS C IO C20 C21 C23 C23 U it -2 9 4 -6 6 14 6 - 2 0 14 - 7 1 20 -1 1 1 22 -1 2 8 22 - 8 1 21 -1 4 1 22 -1 6 5 35 -1 1 0 30 143 33 -2 2 6 38 41 36 -1 1 0 48 120 42 10 35 -1 0 7 40 26 251 - 1 8 34 - 3 5 0 40 313 37 104 16 ») -3391 59 164 66 168 131») Un V im 183(4) 347 7 180 «) 245 16) 231 23 146 24 200 27 253 27 285 26 106 33 541 46 293 41 351 37 293 43 1228 60 168 33 375 35 345 38 478 30 618 53 712 57 745 60 -4 6 2 03 -4 7 1 72 171 6 6 ) 1425 156) -2 8 (4 -9 0 (7 44(6 -4 0 1 -6 4 (2 -3 3 (2 -1 1 7 2 37 2 -5 7 (2 -1 1 3 —191 4 245 4 -7 7 3 —261(( —212(s 303(i 105(4 -1 6 0 : 30: 135 4 —341 4 4io(a 8 C 225 1 221 ! (110) -7 7 (1 H i * fo rm o f th e dleoleceo m t factor is: «p + (fe < V * + VTakkt,’ V + iO u h U 'c' 50 Table VIII. Complete Dietancee and Anglea for Compound 4. Diftance(A) Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Si Si Si Si p p p p -P -N -C l -C 2 -C 3 -C 4 -C 5 -C p * -H S c -H S c ' -N -C l -C ll -C l2 -C IS -C 1 4 -C 1 5 -H S c N -C M C l -C 2 C l -C 5 C2 —C 3 C2 -C 7 C3 -C 4 C3 -C 8 C 4 -C 8 C4 -C 9 C 3 -C IO C M -C 1 7 C M -C M C M -C M C 2 0 -C 2 1 C 2 0 -C 2 2 C 2 1 -C 2 2 C 2 2 -C 2 3 HSc -H S c ' 2.996(1, 2.058(3 2.314 4 2.424 4 2.665 4 2.688 4 2.523 4 2.216 3 1.97 3 1.98 3 1.663 3 1.858 4 1.836 9 1.941 143) 1.812 8 1.786 7 1.847 7 2.93 3 1.478 5 1.471 6 1.423 6 1.381 6 1.523 8 1.459 6 1.543 9 1.400 6 1.4904 8 1.870 8 1.807 9 1.888 14)) 1 .4 M 10) 1.27 2 1.33 2 1.37 2 1.28 2 2.15 4 Anglc(°) -S c -s i P -S c - p N C p * -S e - p HSc -S c - p HSc' -S e - p HSc -S c -N HSc' -S c -N HSc -S c -C p * HSc' -S c -C p * HSc -S c -H S c ' C l -S i - N C l l -S i -N C12 -S i -N C l l -s i - C l C12 -s i - C l C12 -S i - C l l -S e C13 - p C 14 - p -S c C 18 - p -S c C14 - p -C 1 3 -C 1 3 C M -p C 18 - p -C 1 4 -N -S c Si -S c C M -N -S i C M -N C 2 - C l -S i C 8 - C l -S i C 5 - C l -C 2 C 3 -C 2 - C l C 7 -C 2 - C l C 7 -C 2 -C S C 4 -C 3 -C 2 C 8 -C 3 -C 2 C 8 —C 3 -C 4 C 5 -C 4 -C 3 CO -C 4 -C 3 CO -C 4 -C 5 C 4 -C 8 - C l C IO -C 8 - C l C IO -C 8 -C 4 COS' -C 2 0 -C 2 1 C 22 -C 2 1 -C 2 0 C23 -C 2 2 -C 2 0 ' C 23 -C 2 2 -C 2 1 C21 -C 2 2 -C 2 0 ' 108.7(0 95.3(1 112.4(0 68.8(7 134.0 7 125.4 7 104.2 7 129.8 7 102.3 7 65.9 140 97.3 2 113.3 3 116.2 3 110.7 3 112.4 3 106.8 4 124.5 3 115.4 2 117.1 2 96.1 3 90.6(3 90.4(3 101.4(1 135.0(3 123.4 3 122.8 3 121.6 3 100.2 8 107.7 4 120.9 4 121.7 4 106.6 4 121.7 5 130.8 5 111.1 4 126.3 4 122.0 4 105.3 4 120.4 4 125.1 4 114.6 14 124.2 14 100.7 15 120.3 15 121.0 14 C19 Figure 1.8. ORTEP drawing of [(Cp‘SINR)(PMS3 )ScH]2 (4) showing thaatomlabollng. 52 Table IX. Crystal and Instonsity Collection Data for [(Cp*SINR)Sc]2 (M 2,n2-C2 H4) (5). Formula: Sc2 Si2 P 2 N 2 C4 sHg4 Formula Weight: 861.22 Crystal Color pale yellow Habit: Hexagonal Prisms a =23.706(5) A 6 = 11.342(2) A 8 = 111.57(1)* c=21.141(3)A v = 5286.2(7) A3 z=4 X=0.71073 A T: 23* Graphite monochromator Space group: C2/c Absences: hkl, h + k odd; hOl, 1odd Crystal Site: - .35 z .4 z.4 mm u=4.03cm*t (lirma*=0.13) CAD*4 Diffractometer goscan 28 range: 2*-40* Octants collected: ± h , ± k ,-l Number reflections measured: 5205 Number of independent reflections: 2475 Number with F«2>0: 2291 Number with F«2>3o(V«8): 1757 Goodness offlt fee merging data: 0.97 Final R-indes: 0.086 Final goodness of flt: 3.84 53 Table X. Final Parameters for Compound 5. x , y , z and x 104 X y z Sc 1008(1) 485(1) 2979(1) 452(4) P 1128(1) 157(2) 4347(1) 729(8) Si 2091(1) -3 8 0 (2 ) 2729(1) 053(7) N 1478(2) -0 8 8 (5 ) 2803(3) 4.2 (1 ) • C 1 1800(3) 1190(0) 2751(4) 4 .1 (2 ) • C 2 1301(3) 1717(0) 2237(4) 4 .3 (2 ) • C 3 1114(3) 2007(5) 2520(4) 4 .4 (2 ) . C 4 1449(3) 2535(7) 3213(4) 4 .0 (2 ) • C 5 1905(3) 1803(7) 3358(4) 4 .7 (2 ) « C 6 1132(4) 1412(7) 1489(4) 789(27) C 7 030(4) 3457(8) 2129(5) 953(30) C 8 1379(4) 3554(8) 3587(5) 977(30) C 9 2414(4) 1550(8) 4052(4) 894(30) C IO 2 8(4) 455(9) 2174(4) 4 .3 (2 ) C ll 2174(5) -7 5 5 (8 ) 1914(5) 1142(34) c is 2S57(4) -7 2 0 (8 ) 3374(5) 1158(37) C IS 1349(4) -2 2 5 5 (8 ) 2845(5) 744(29) CM 848(5) -2 5 1 3 (8 ) 3054(5) 1283(41) C IS 1854(5) -2 9 4 0 (1 1 ) 3234(11) 2730(102) C IS 1129(10) -2 5 5 4 (1 1 ) 2124(10) 3085(105) C 17 1019(5) 1443(10) 4784(5) 1218(38) C IS 553(4) -7 5 7 (9 ) 4455(4) 1044(33) C 19 1810(4) -4 8 2 (1 1 ) 4955(5) 1424(43) A tom Utq or B • 54 Table X. (Cont’d) H10A -7(28) 1130(58) 3045(32) 0.2(20) • H10B -91(23) -234(49) 3010(27) 3.8(15) . ‘ Isotropic displacem ent param eter, B 55 Table XI. Assigned Parameters for Compound 5. x, y, z and x 104 A tom X y z C21 C22 C23 C24 H 6A HOB HOC H 7A H 7B H 7C H 8A 1 H 8A 2 H 8A 3 H8B1 H8B2 H 8B 3 H 9A HOB HOC H 11A H11B H llC H 12A H 12B H12C H 14A H 14B H 14C H 15A H 18B H 15C H 10A H 16B H 16C H 17A H 17B H 17C H 18A H 18B H 18C H 19A H 19B H 19C 5008 5543 5536 6124 795 994 1446 433 301 779 1641 1464 965 1731 1036 1303 2804 2480 2339 1795 2478 2283 3148 2993 2847 960 705 504 2218 1801 2016 758 1405 1240 1394 1003 098 577 611 154 2167 1900 1790 829 530 -3 0 0 -6 2 6 1908 603 1514 3851 3049 4054 4245 3287 3881 3055 3414 4345 1953 843 2088 -1 0 5 0 -1 3 4 3 -0 6 -1 5 3 -1 4 9 0 -7 2 8 -2 4 3 0 —3335 -2 0 1 9 -2 7 0 6 -3 7 8 3 -2 8 0 1 -3 8 8 9 -3 3 7 8 -2 0 9 0 1979 1209 1916 -7 7 8 -1 0 0 3 -5 5 0 39 -1 2 2 0 -4 9 0 487 399 -8 8 -1 8 4 1217 1391 1291 2382 1746 1891 3715 4136 3519 4091 3817 3462 4059 4208 4419 1586 1971 1715 3360 3294 3827 3571 2978 2871 3078 3153 3705 1993 2152 1883 4900 5220 4629 4920 4352 4174 5000 4840 5410 • - i £, U,i or 12.0 12.0 12.0 12.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 • «)1 _ * Isotropic displacement parameter, B 56 Table XII. Anisotropic Displacement Parameters for Compound 5. Atom Uu U22 Sc P 428(9) 653(16 512(15 802(64 849(70 1311(84 714 64) ' 1505 93) 655 65) 788 68) 1248 92) 987 105 6416 409 I 6661 106 1052i 77) 1027 :« i) 504 10 1032 21 692 18 979 72 847i 69 675 65 1157 80 834! 76 1074 82 533 65 726 75 991(104 979 112 1348 M ) 1391 93) 2288 130) Si C 6 C 7 C 8 C 9 C ll C12 C13 C14 C IS CIO C17 C IO C IO i 440 ») 484 15 825 18 611 60 1268i 84 1248< 86 745 67 1546 100 1716 103' 1014 76) 2279 134 5807 361 3070 227 821 77 670(63) 839(74) U \2 V iz -5 0 (8 ) - 8 5 15) 5(14) 177(7) 188(12) 327(13) -200 97 -2 9 1 -2 8 6 -1 1 8 107 40 -1 7 0 139 -1 9 9 2 -1 5 2 -2 8 0 56 669(76 293(57 203(66 U23 -2 4 (8 ) 55(14 —73(15 28(52 270(61 -3 3 4 (5 9 -8 1 (5 6 -2 4 3 (6 6 ) -5 4 (7 5 76(56) 41(77) 1410(160 -1 0 73 (1 3 4 -2 1 9 (6 8 169(60 597(82 T h e form o f the displacem ent factor is: exp - 2 jr 3(£7nA, a ** + C733Jk36 '* + C7M / 3c '* + 2 UX2 hka*b* + 2 ViShlo*c' + 2Un kibuf ) 57 Table XIII. Complete Distances and Angles for Compound 5. Diatance(A) Sc -P Se -N Sc -C 1 Se -C 2 Sc -C 3 Sc -C 4 Sc -C 5 Se -C IO Sc -C IO ’ C IO -C IO ’ C IO -H lO A C IO -H lO B Si - N St - C 1 Si - C l l Si -C IS C l l -H 1 1 A C l l -H U B C l l -H 1 1 C C IS -H 1 S A C IS -H 1 2 B C IS -H 1 S C N -C IS C IS -C 1 4 C IS -C IS C IS -C IO C14 -H 1 4 A C14 -H U B C14 -H 1 4 C C IS -H lS A C IS -H lS B C IS -H 1 S C C IO -H lO A C IO -H lO B C IO -H IO C P - C lT P -C IS P -C 1 9 C l -C S C l -C S C S -C S C S -C O CS -0 4 C S -C T C 4 -C S C 4 -C O C S -C O 2.825(3) 2.071(6) 2.305(7) 2.464 8) 2.641 8) 2.628 ,8) 2.481 8) 2.320 9) 2.357 «) 1.433 12) 0.88 7) 0.03 0 1.722 0) 1.871 o) 1.8SS 11 1.865 10 0.007 10 0.048 10 0.970 10 0.0S3 10 0.008 10 0.007 10 1.400< 11 1.4SS 15 1.41 2) 1.40 2) 1.007 H] 0.0054 11 0.947 U 1.0S7 17 0.0731 17 0.989! 17 0 J 7 3) I M Si 0.91! 3 ) 1.790! W] 1.000! 10 1.007 11 1.419! 11 1.450! 11 1.4051 11 1.513 11 1.309! 11 1.493! 13 1.301 11 1.505 13 1.513 IS 1 Dietanct(A) C 0 - H OA C 0 - H OB C 0 - H OC C 7 -H 7A C 7 -H 7B C 7 - H 7C C 8 -H 8 A 1 C 8 -H 8 A 2 C 8 -H 8 A 3 C 8 -H 8 B 1 C 8 -H 8 B 3 C 8 -H 8 B 3 C 9 - H OA C 9 - H OB C O -H O C C 17 -H 1 7 A C 17 -H 1 7 B C 17 -H 1 7 C C 18 -H lO A C 18 -H lO B C IO -H 1 0 C C IO -H lO A C 19 -H lO B C 1 9 -H 1 9 C C31 -C S S C 3 1 -C 3 3 CSS -C S S C 3 3 -C 3 4 0.074(8) 0.971(8) 0.985(8) 0.943(9) 1.007 r9 0.983 ,9 0.980 9 0.947 9 0.082 9 0.055 9 0.903 9 0.900! 9 0.978! 9 0.077 9 0.984 9 1.038 11 0.057 11 0.9S3 11 0.955 9 0.999 0 0.940 9 1.005 11 0.935! 11 0.970! 11 1.390! 0 1.391 0 1.390! 0 1.520! 0 Anglt(a) CIO -Sc -C IO ’ -C IO ’ 73.6(5 35.7/3 C14 -C IS C IS -C IS C IO -C IS C IS -C IS C IO -C IS C IO -C IS CSS -C 2 1 CSS -css css-css C24 -CSS C 2 4 -C 3 S CIS -IV -Oft 124.41(5) [(CpaSINRHPM»a)ScMi,i)V-P2H4) (5) Figure 1.9. OfiTEP drawing ol 913 59 Table XIV. Crystal and Intensity Collection Data for [(Cp*SINR)Sc]2 (M*CH2CH2 CH3) (6). Crystal Color: pale yellow Crystal Habit: irregular plates a =9.429(2)A 6=21.937(5)A 6=99.39(2)* c = 9.826 (2)A T = 21*C X= 0.71073 A CAD-4 20 range: 3*- 40* ascan Octants ofdata collected: ±h , ± k , 1 Graphite monochromator: yet p = 4.28cm*l prmax =0.16 Crystal Sim: .2 a . 2 x . 8mm Total Number Reflections: 4082 Total Independent Reflections: 1863 GOF for Merging: 0.977 (R merge for reflections with exactly 2 observations) 0.041 Number of Reflections used in Refinement: 1863 Number ofReflections with F **> 0 used in R: 1708 Number of Reflections with F o *> 3 o (F # *): 1231 (R for reflections with F0, >3o<Fo>): 0.0579 60 Tabla XV. Final Paramatara for Compound 6. x, y, z and x 104 z y z Se -8 5 (2 ) 748(1) 5102(1) 460(4) Cl 464(8) 1783(3) 5817(8) 463(21) C2 1742(0) 1557(3) 5417(8) 503(24) C3 2422(8) 1151(3) 6410(0) 522(25) C4 1560(10) 1115(3) 7452(8) 558(26) CS 368(8) 1400(3) 7083(9) 521(25) CT 2304(8) 1745(3) 4169(0) 868(27) CS 3020(0) 912(3) 6481(10) 087(32) CO 1052(0) 700(4) 8863(8) 074(31) C IO -7 4 7 (0 ) 1620(4) 7905(0) 878(27) Si -1 1 0 8 (2 ) 2023(1) 4535(2) 507(7) C ll -2 5 2 7 (8 ) 2423(4) 5348(9) 1070(31) C 12 -6 0 0 (0 ) 2504(3) 3202(10) 1046(32) N -1 5 3 7 (5 ) 1300(2) 3023(5) 419(16) C IS —2084(8) 1166(3) 2711(8) 556(26) C 14 -3 2 5 3 (1 0 ) 537(4) 2868(10) 730(33) C IS -3 0 3 9 (0 ) 1615(4) 2611(10) 608(31) C IS -1 0 0 5 (1 1 ) 1205(5) 1386(10) 813(41) C 30 885(8) 111(3) 3677(8) 501(23) C 21 2222(10) -1 7 3 (4 ) 3307(0) 920(31) C 22 2646(0) -6 4 (4 ) 1071(0) 1080(33) A to m • a * - i E i U.n • % )1 61 Table XVI. Assigned Parameters for Compound 6. x , y end x x 104 A to m X y X B C 14» C 15» C 16* H 7a H 7b H7c H 8a H 8b HSc HO* HOb HOe H lO a H lO b H lO c H lla H llb H lle H 12a H l2 b H12« H id e H 14b H id e H it e H IB b H IS e H it e H l« b H it e H S te H 30b H lle H 31b H 33a H 33b H33e -4 1 0 1 —227d -2 8 5 2 1718 3230 26d7 4050 4101 4587 2526 2400 1103 -5 3 6 -6 8 3 -1 6 5 7 -2 0 5 3 -3 0 8 3 -3 1 1 0 -1 4 7 3 -2 0 1 40 -3 6 8 0 -2 4 3 0 -3 8 7 7 -4 3 1 7 -4 6 1 7 -3 5 8 3 -1 0 7 4 -2 6 8 7 -1 1 3 9 1110 189 2048 2808 1070 3569 2701 1054 507 1714 1668 1400 2158 760 600 1242 1070 448 705 1008 1209 1619 2603 2134 2659 2810 2400 2883 516 270 458 1594 1473 1006 1638 1000 1010 546 37 -5 8 1 -4 -2 5 3 -2 4 6 361 3277 1080 1752 3374 4161 4262 5614 7167 6746 0474 8742 0203 8433 8665 7437 6045 5721 4648 2803 2610 3703 3613 2018 2011 3433 1838 2420 1178 703 1553 3653 2835 3543 4103 1305 2033 1833 10.3 10.3 10.3 8.0 8.0 8.0 8.0 8 .0 8 .0 0 .6 0 .6 0 .6 8.3 8.2 8.2 10.1 10.1 10.1 0.8 0 .8 0.8 12.1 12.1 12.1 11.3 11.3 11.3 13.3 13.3 13.3 5 .6 5.6 8.7 8.7 0 .9 0 .0 0 .0 62 Table XVII. Anisotropic Displacement Parameters for Compound 6. A to m Un Un I7m Se Cl C2 C3 C4 C5 C7 CS CO C IO Si C ll C 12 N C IS C 14 C IS C IS C 20 C 21 C 22 507 10 564 56 593 60 458 57 722 65 660 64 878 66 685 63 1445 83 1056 73 684 17 1017 72 1245 75 403 41 685 59 819 76 547 67 010 83 545 51 1064 78 054 [60 332 8) 371 46] 340 48 385 51 302 51 323 48 712 58 640 57 706 50 773 60 346 12 875 64 528 53 200 32 338 50 452 63 532 60 1197 02 379 46 003 76 1604 91 440(10 478(55 562 60 670 64 490 61 606 64 1081(78 15321 80 627 57 841 68 730 17 1244 86 1237 81 471 30 572 50 804 70 005< 81 324 65 856 64 678 60 740 69 Uu 22 8 ), 10 40) -5 5 44 60 44 -0 1 48 - 5 6 45 -1 0 1 >® 28 50 45 5® - 3 2 55 68 12 437 54 -1 4 50 74 28 16 45 29 55 80 55 -1431 60 -2 1 41 -1 0 4 50 264 67 Uu - 0 8) 154 44 82 52 - 6 7 53) -1 0 0 (5 5 183(54 357(61 -1 0 1 62 -2 6 2 58 265< 61 23 -1 1 2 66 -1 8 1 66 74 34 - 1 1 4 50 -2 8 7 66 -2 1 0 60 8 4 61 134 48 77 61* 316 58 Un 19 8) 56 42) -3 8 i 44) -7 7 47) - 2 4 46) -2 0 8 46 - 1 0 i56 i -2 3 7 (5 8 60(54 -2 1 8 (5 3 50 i12 - i -3 4 1 61 470 i56i 56 29 10 44 - 4 55 - 4 1 50 - 8 5 62 -1 8 4 43 42(56 355(65 Tfca form of the displacement facto r is: « p -2»*(ITMA V * + Un kH'* + + 3CTi3M s * » * + 2Uit hU 'e' + 2Un kik'cu) 63 Table XVIII. Complete Distances and Angles for Compound 6. Distance(A) Sc -N Sc -C 2 0 Sc -C 2 0 Sc -Sc C20 -C 2 0 Sc -C p * Si -N SI —C l Si - O i l Si -C 1 2 C20 -C 2 1 C21 -C 2 2 C l -C 2 C l -C 3 C 2 -C 3 C 2 -C 7 C 3 -C 4 C 3 -C S C 4 -C S C 4 -C O CS -C IO N -C IS C IS -C 1 4 C IS -C IS C IS -C IS C IS -C 1 4 A C IS -C 1 S A C IS -C 1 6 A 2.083(5 2.334(7 2.372 7) 3.310 2) 3.340 10) 2.210 1.720 5 ) 1.858 7) 1.885 8) 1.862 S) 1.472 11 1.538 12 1.417 10 1.407 10 1.403 11 1.515 11 1.403 11 1.507 11 1.403 11 1.5391 11 1.513 11 1.501 Q) 1.407 11 1.530 11 1.550 12 1.548 7 1.521 7 1.3201 7 AafteC) CO - C 4 - C 5 C 4 -C S - C l C IO -C S - C l C 1 0 -C S -C 4 C ll-S i - C l C 1 2 -S I - C l N -S i - C l 0 2 -S i - O i l N -S i - C l l N -S i -C 1 2 C IS - N -S i 103.7 115.5 1 1 7 .9 123.5 8« 90 132.8(4) 103.4(2) •IV •» -CIS -SI 124.61 109.31 126.61 1 23.9 112.8 112.1 9S .3 A n g le(°) -C 2 0 -C 2 1 -C 2 0 -S c -C 2 0 -C 2 1 -S c -C p * -S c -C 2 0 -C 2 1 -C 2 0 -S c -C 2 0 -S c -C 2 0 - C l -S i -C p * - C l - C l -C 2 -C 2 - C l -C 2 - C l -C 2 -C S -C S -C 2 CS -C S -C 2 CS -C S -C 4 CS —C 4 -C S CO -C 4 -C S CO -C 4 -C S C4 -C S - C l C IO -C S - C l C IO -C S -C 4 C 14 - O S - N C IS - O S - N C 16 -C IS - N C 1 4 A -C 1 3 - I f C lS A -C lS - N C 1 6 A -C 1 3 - N C1S - C IS -C 1 4 C 1 6 - C IS - 0 4 C 16 - O S - O S C lS A -C lS -C 1 4 A C lS A -C lS -C 1 4 A C lS A -C lS -C 1 5 A CSS -C 2 1 -C 2 0 CS - C l -C 2 Si - C l -C 2 Si - C l -C S CS -C 2 - C l C7 -C 2 - C l C7 -C 2 -C S C4 -C S -C 2 CS -C S - 0 2 CS -C S -C 4 CS -C 4 -C S CO -C 4 -C S Sc Sc Sc N C20 C22 C p* C p* C p* Sc C5 CS C7 C7 C4 144.0(5) 80.4 2 95.7 5) 102.5 00.6(2) 117.5 7 110.6 118.5 155.6 82.8 105.7 6) 100.8 7 127.3 7 122.8 7 106.9 7 124.7 7 127.3 7 108.3 7 126.5 7 124.6 7 100.3 7 126.6 7 123.9 7 108.0 6 112.0 6 108.0 6 107.4 5 100.2 5 100.0 5 106.3 6 110.0 7 100.6 6 100J 4 100.9 4 111.4 4 117.5 7 105.7 6 122.1 5 124.5 5 100.8 7 127.3 7 122.8 7 106.9 7 124.7 7 127.3 * 108.3 7l 126.5 * 64 65 Table XIX. Crystal and Intansity Collection Oata for (Cp*SINR)(PMt3)ScCH2TMS (16). a = 11.181(4)A Crystal Color: clear 6=18.206<5)A Crystal Habit: irregular c= 15.250 ( 6 )A 29 range: 2*- 40* B= 108.25 (3)* Octants of data collected: ± h, ± k, I X= li= 3.95cm* i 0.71073 A T = 23*C CAD-4 <uscan Graphite monochromator: yea pTmo a 0.15 Absences: h 0 l f h + lodd;0k0,kodd. Crystal Size: 0.14 a 0.26 a 0.67mm Total Number Reflections: 8506 Total Independent Reflections: 5186 GOF for Merging: 0.94 (R merge for reflections with saactly 2 observations: 0.039) Number of Reflections used in Refinement: 5186 Number of Reflections with Fo*> 0 used in R: 4307 Number of Reflections with F0*>3o(Fo*): 2101 (Rfbr reflections with F0s>3o<Fot : 0.0533) 66 Table XX. Final Parameter* for Compound 16. f • x 104 * .y . x and U% A tom x y X Sc 5569(.6) 1832(.4) 2742(.5) 489(2) P 5936(1) 1887(1) 1031(1) 734(4) S il 7502(1) 948(1) 4101(1) 712(4) Si2 2285(1) 1868(1) 2735(1) 769(4) N 6133(3) 794(2) 3196(2) 567(9) Cl 7460(4) 1960(2) 4019(3) 589(12) C2 6349(4) 2416(2) 4246(3) 586(12) C3 6296(4) 3048(2) 3692(3) 571(12) C4 7017(4) 2983(2) 3101(3) 652(13) CS 7716(3) 2345(3) 3299(3) 616(13) C6 3494(3) 2056(2) 2189(3) 661(12) C7 481 1 (8 ) 1427(3) 7 1 (3 ) 1172(20) CS 5828(5) 2806(3) 586(3) 1325(22) C9 7428(5) 1566(4) 955(4) 1880(30) C IO 8 983(4) 6 03(3) 3912(4) 1079(17) C ll 7515(5) 657(3) 5277(4) 1242(20) C12 1261(5) 1074(3) 2170(4) 1304(20) C IS 2980(5) 1619(3) 3994(4) 1135(17) C 14 1230(5) 2670(3) 2686(4) 1322(20) C IS 5536(5) 3718(2) 3792(4) 811(16) 67 TablaXX. (Cont’d) A tom X y z C16 4571(5) 3542(3) 4261(4) 1222(20) C17 4862(6) 4058(3) 2846(5) 1428(24) C18 6456(5) 4297(3) 4371(4) 1271(21) C19 5596(5) 66(2) 2916(3) 751(15) C20 4921(10) -2 1 3 (4 ) 3537(6) 2722(42) C21 6444(6) -4 7 5 (4 ) 2802(8) 2933(48) C22 4577(8) 123(3) 2026(6) 2163(38) 68 Table XXI. Assigned Hydrogsn Parameters for Compound 16. x , y and * x 104 A to m H2 H4 H5 H 6* H 6b H 7s H7b H7e H 8* H 8b H8c H 9e H flb HOe H lO e H lO b H lO c H lls H llb H llc H 12* H l2 b H12e H lS s H 13b H13e H 14s H 14b H id e H 16s H lfb H l6 e HX7s H 17b H l7 c H lS s H IS b H lS e H 20s X y z 3 6178 7030 8296 3221 3438 5081 4801 4007 5984 5021 6458 7427 8079 7537 2307 3325 2188 1798 2570 1495 922 1637 2790 2992 3098 1619 1859 1072 711 88 837 769 909 142 995 1188 850 1527 1192 2014 2542 3070 2793 3979 3187 3358 4714 2632 2996 1819 2072 -4 8 5 207 -2 3 0 522 998 336 1349 1137 4442 3818 3388 5705 5434 5272 2478 1545 2228 4239 4044 4310 2980 2999 2083 4302 3907 4859 2938 2585 2477 4433 4957 4083 3343 5.5 6.1 5.7 6.1 6.1 11.0 11.0 11.0 12.3 12.3 12.3 17.9 17.9 17.9 10.1 10.1 10.1 11.5 11.5 11.5 12.3 12.3 12.3 10.5 10.5 10.5 12.4 12.4 12.4 11.5 11.5 11.5 13.5 13.5 13.5 12.3 12.3 12.3 26.0 9872 8913 9084 8313 887S 7374 653 882 1778 2308 3478 3474 8 30 1737 827 4127 4004 4984 4398 5483 4304 5984 6877 7038 4832 4473 4203 3708 4718 4099 4427 —697 69 Table XXI. (Cont’d) Atom H20b H20c H 21a H21b H21c H 22a H22b H22c B y 5555 4303 5992 6777 7091 4253 3957 4931 -2 6 4 96 -9 1 9 -3 1 5 -5 3 7 -3 5 9 435 302 4151 3563 2619 2334 3363 1861 2099 1581 26.0 26.0 28.7 28.7 28.7 20.5 20.5 20.5 70 Table XXII. Anisotropic Oitplacomont Parameters for Compound 16. A to m U \ u £T.i i N 507(5) 1105(10) 607(9) 717(9) 48 Cl C2 C3 C4 CS C6 C7 CS C9 707(31 1412(48 1490(52 3950(101) Se P 511 512 138(7) CIO C ll C12 C IS C14 C IS CIO C17 C IO C IO C20 C21 C22 847 41 2407 94 Uu Taleee have ben aaldpUed by 10* Xh* fo n t a i the d iip lu eam t betor in exp -2 **(lT „ b *n *' + Un * * ' + + 2C7„U**e*) t 'u 5 (8 ) 8(17) -9 (2 5 ) -5 8 24 - 5 8 23 -1 2 8 20 - 7 0 24 -1 8 22 4 38 -3 1 4 44 900 54 3001 29 28 39 -4 1 8 37 3 8 34 3041 3 5) - 2 8 30) 9 8 3 3) 4 09 40 -1 0 0 34 - 2 5 29 •2143 39 40 43 -0 7 8 53 + 2 P , 34 4- 2t7«Me*e* 30(4) 67(8) 109(7) 1(9) 0(18) 21(25) -42(25) -25(25) 71 25) -93(27) 9(24) -101(34) 262(37) 127(56) 27(34) 259(33) -128(43) 93(34) 277(44) —31(29) -372(40) 156(41) -455(39) -71(29) -929(64) -2150(85) -250(52) 71 Tablt XXIII. Complata Diatancaa and Anglas for Compound 16. Distmnce(A) Sc -C p Sc -P Sc -N Sc - C l Sc Sc Sc Sc Sc Sc Sc Sc -C 2 -C 3 -C 4 -C 5 —CS -a i3* -H l3 b -H l3 c -C 7 —CS P P P -c» Sil -N Sil - C l sil -C IO Sil - C l l Si2 “ CS Si2 -C IS sia -C IS Si2 -C 1 4 N -C IS Cl -CS Cl -C S C2 -C S C2 -BS CS -C4 CS - C IS C4 -C S C4 -B4 CS -as CS - B t e CS -H S b C 7 -B 7 a C 7 -B 7 b C 7 -H 7 e CS -B S a CS -B S b 2.208 2.767(1) 2.044(3) 2.339(4) 2.454(4) 2.632(4) 2.600(4) 2.467(4) 2.244(4) 4.901 3.701 3.851 1.310(5) 1.796(6) 1.804(7) 1.731(3) 1.846(4) 1.876(5) 1.866(5) 1.829(4) 1.876(6) 1.867(5) 1.864(6) 1.464(6) 1 .431(6) 1.410(6) 1.403(6) 0.951 1.390(6) 1.521(6) 1 .379(6) 0.952 0.951 0.951 0.950 0.950 0.943 0.945 0.952 0.940 DUtance(A) CS C9 -H 8 c -H 9 * C9 -H 9 b C 9 -H 9 c C IO -H lO a C IO -H lO b C IO -H lO c C ll -H 1 U C ll - H llb C ll -H llc C 12 -H 1 2 * C 12 -H l2 b C12 -H 1 2 e C13 -H 1 3 * C IS -H 1 3 b C IS -H 1 3 e C 14 - H l t o C 14 -H 1 4 b C 14 -H l4 c C 15 -C IS C 15 -C 1 7 C 15 -C IS C IS - B IS * C IS -B 1 6 b C IS -B 1 6 c C 17 -B 1 7 a C 17 -H 1 7 b C 17 -B 1 7 e C IS -B lS a C IS -B IS b C IS -H IS e C 19 -C 2 0 C 19 -C 2 1 C 1 9 -C 2 2 C 2 0 -B 2 0 * C 20 -B 2 0 b C 20 -H 2 0 e C21 - H 2 U C21 -H 2 1 b C21 -H 2 1 e 0.944 0.949 0.950 0.939 0.946 0.949 0.943 0.948 0.942 0.950 0.951 0.939 0.949 0.952 0.945 0.942 0.944 0.949 0.944 1.504(8) 1.533(8) 1.543(7) 0.951 0.946 0.944 0.952 0.941 0.947 0.952 0.942 0.945 1.472(10) 1.415(9) 1.478(9) 0.95S 0.98S 0.901 0.947 0.948 0.939 72 Tabl* XXIII. (Cont’d) A ngle(“ ) 9 6 * -CO —Si2 9 6 b -C 6 -S i2 H6b -C 0 - 9 6 * 9 7 s -C 7 -P 9 7 b -C 7 -P 9 7 c -C 7 -P 9 7 b -C 7 - 9 7 * 9 7 c -C 7 - 9 7 * 9 7 c -C 7 -9 7 b 9 8 * -C 8 -P 9 8 b -C 8 -P 9 8 c -C 8 -P 9 8 b -CO - 9 8 * 9 8 e -CO - 9 8 * 9 8 c -C 8 -9 8 b 9 9 * -C 9 -P 9 9 b -C 9 -P 9 9 c -C 9 - P 9 9 b -C 9 - 9 9 * 9 9 c -C 9 - 9 9 * 9 9 c -C 9 -9 9 b 9 1 0 * -C IO -S IX 9 1 0 b -C IO - S il 9 1 0 c -C IO - S il 9 1 0 b -C IO -B IO * H lO c -C IO - 9 1 0 * H lO c -C IO -H lO b 9 1 1 * - C l l -S il 911b - C l l - S I B ile - C l l - S il B llb - C l l - B l l * B ile - C l l - B l l * B ile - C l l - B l l b B 1 2 * -C IS -S IS B lS b - C lS - S ia B l2 e - C IS -S IS 9 1 2 b - C IS - B IS * B IS c - C IS - B IS * B IS c - C IS -B IS b 9 1 3 * -C IS -S IS A ngle(#) 104.6 104.7 109.4 108.6 109.0 108.8 110.1 109.8 110.8 108.3 108.9 108.7 110.2 109.8 110.8 108.8 108.3 109.1 109.8 110.5 110.5 108.8 108.8 108.8 109.9 110.4 110.1 108.8 109.2 108.9 110.3 109.9 110.1 108.9 109.2 108.7 110.3 109.5 110.5 108.5 9 13 b -C 1 3 -S i2 913c -C 1 3 —Si2 913b -C 1 3 - 9 1 3 * 913c -C 1 3 -H 1 3 * 913c -C 1 3 -9 1 3 b 9 1 4 * -C 1 4 —SI2 9 14 b -C 1 4 -S i2 914c -C 1 4 -S i2 914b - 0 4 -9 1 4 * 9 14 c -C 1 4 - 9 1 4 * B14c - 0 4 -9 1 4 b C 18 —0 5 —C 3 C 17 -C IS -C 3 C18 -C IS -C 3 C 17 -C 1 5 -C 1 8 C 18 - O S -C 1 8 C 18 - O S —0 7 9 1 8 * -C 1 8 -C IS B 18b -C 1 8 -C IS 9 1 6 c -C 1 6 -C IS B 16b -C 1 6 - 9 1 8 * 9 1 6 c -C 1 6 - 9 1 8 * B18c - 0 6 -9 1 8 b B 1 7 * -C 1 7 -C IS 9 1 7 b -C 1 7 - O S B17c -C 1 7 - O S B 17b -C 1 7 -B 1 7 * B l7 c - 0 7 -B 1 7 * H17e -C 1 7 -B 1 7 b B IS * -C 1 8 - O S B 18b -C 1 8 - O S B l8 c -C 1 8 -C IS B IS b -C 1 8 -B 1 8 * 9 1 8 c -C IS -B 1 8 * B IS c -C IS -B IS b - S il C 19 - N C 20 -C 1 9 - 9 C21 - C 1 9 - 9 CSS - 0 9 - 9 C21 -C 1 9 -C 2 0 108.8 109.1 109.8 110.0 110.7 108.8 108.5 108.8 110.0 110.5 110.1 112.5(4) 110.8(4) 108.1(4) 108.6(4) 108.8(4) 107.8(4) 108.8 109.1 109.1 109.8 109.9 110.3 108.3 109.3 108.8 110.1 109.8 110.8 108.4 109.1 109.1 110.0 109.7 110.8 124.0(3) 111.8(5) 115.8(5) 109.4(4) 109.8(8) 73 Tabl* XXIII. (Cont’d) aaee(A ) C22 -H22a C22 -H22b C22 -H22c 0.952 0.930 0.943 Angle(#) Cp -S c -N Cp -S c -P Cp -S c -C fl Sc -P -C 7 Sc -P -C 8 Sc -P -C 9 C8 -P -C 7 C9 - P -C 7 C9 -P -C S Se - N -C 1 9 Sc - N - S il Sc -C S -S iS C l - S il - N C IO - S i l - N C l l -S il -N C IO - S i l - C l C l l -S il - C l C l l - S i l -C IO C IS -S i2 -C S C IS -S i3 -C S C 14 -S i2 -C S C IS -S IS -C IS C 14 -S iS -C IS C 14 -S iS -C IS S il - C l -C p CS - C l - S il CS - C l - S il CS - C l -C S CS -C S - C l H 3 -C S - C l H 3 -C S -C S C 4 -C S -C S C IS -C S -C S C IS -C S -C 4 CS - C 4 - C 3 H4 -C 4 -C 3 H4 -C 4 -C S C 4 - C S -C 1 HS —CS —C l H5 -C S -C 4 104.1 106.9 126.3 118.4(2) 112.2(2) 117.5(2) 100.2(2) 102.8(3) 103.3(3) 133.1(3) 102.9(2) 128.3(2) 9 6.1 (2 ) 115.2(2) 117.3(2) 109.5(2) 109.8(2) 108.1(2) 111.5(2) 112.4(2) 112.5(2) 106.3(2) 107.3(3) 106.4(2) 154.1 124.9(3) 122.6(3) 103.7(3) 110.8(4) 124.4 124.8 105.8(4) 127.7(4) 126.1(4) 109.5(4) 125.4 12S.1 1 10.2(4) 124.8 125.0 74 Table XXIII. (Cont'd) Angle(°) C22 -C 1 9 -C 2 0 C22 -C 1 9 -C 2 1 H20a -C 2 0 -C 1 9 H20b -C 2 0 -C 1 9 H20c -C 2 0 -C 1 9 H20b -C 2 0 -H 2 0 a H20c -C 2 0 -H 2 0 e H20c -C 2 0 -S 2 0 b H21» -C 2 1 - C l 9 H21b -C 2 1 -C 1 9 H21c -C 2 1 -C 1 9 H21b -C 2 1 - H 2 U H21c -C 2 1 -H 2 1 * H21c -C 2 1 -H 2 1 b H 2 2 * -C 2 2 -C 1 9 H22b -C 2 2 - C l 9 H22e -C 2 2 -C 1 9 H22b -C 2 2 -H 2 2 e H22e -C 2 2 -H 2 2 a H22c -C 2 2 -H 2 2 b 102.3(5) 106.8(5) 108.1 106.4 111.6 106.3 113.3 110.5 108.3 108.3 109.0 109.9 110.7 110.6 107.3 108.8 108.0 111.0 109.8 111.8 70 EXPERIMENTAL SECTION General Considerations. All manipulations were performed by using glovebox and high-vacuum line techniques as described elsewhere. 2 6 Solvents were dried by standard techniques and further purified by vacuum transfer from titanocene2 7 or sodium benzophenone. Argon and hydrogen were purified by passage over MnO on vermiculite and activated 4A molecular sieves. NMR spectra were recorded on Varian EM-390 (1 H, 90MHz), JEOL FX90Q (1 H, 89.56 MHz; 1 3 C, 22.50 MHz; 3 1 P, 36.23 MHz), JEOL GX400Q (1 H, 399.78 MHz;1 3 C, 100.38 MHz) and Bruker AM500 ^H, 500.13 MHz) spectrometers. Infrared spectra were recorded on a Beckman 4240 spectrometer and a Perkin-Elmer 1600 series FTIR spectrometer and peak positions are reported in cm'1. Elemental analyses were performed by F. Harvey of the Caltech Analytical Laboratory. Where indicated, oxidant (V2 OS and WO 3 ) was added to the compound to facilitate burning. Reagents. Propene, ethylene and HCI (Matheson) were freeze-pump-thawed at least twice prior to use. PMea was either prepared as described elsewhere2 6 or purchased from Aldrich. Butyllithium (1.6M soln. in hexanes) and M e^iCfe were used as obtained from Aldrich. Methyllithium and trimethylsilylmethyllithium (Aldrich) were used as solids obtained by drying their ether solutions. ScCtofTHFJa,1** U(CsMe4 H) , 7 UCH(SiMe3 )2 , 2 6 and UCH 2 PMe2 3 0 were prepared as reported elsewhere. U 2 (C p* 8 INR) (1). U(C 5 Me 4 H) (23.1 g, 0.2 mol) was suspended in 250mL tetrahydrofuran. While cooling the mixture to -78*C, Me2 SiCl2 (24.1 mL, 0.2 mol) was added via syringe. The mixture was allowed to warm to room temperature and stirred overnight The tetrahydrofuran was removed in vacuo, leaving (CsMe4 )SIMe2 CI as a non-volatile, clear, colorless liquid, which was dissolved in 100 mL petroleum ether and filtered to remove LiCI. The petroleum ether was removed in vacuo, the (CsMe4 )SiMe2 CI was dissolved in 250 mL 77 THF and Li(Me3 CNH) powder was added gradually with stirring to the solution at room temperature (exothermic). The mixture was stirred for two hours at room temperature, and the THF was removed in vacuo and replaced with 100 mL petroleum ether. LiCI was filtered off and the petroleum ether was removed in vacuo to afford H2 Cp*SiNR as an orange oil. The oil was dissolved in 100 mL diethyl ether and cooled to -78°C. n-Butyllithium ( 1 .6 M in hexanes, 250 mL, 0.4 mol) was added via syringe and the reaction was allowed to warm to room temperature. After stirring the reaction overnight, the precipitated Li2 (Cp*SiNR) was collected by filtration as a tan powder (45.2 g, 80.7% yield). (Cp*SINR)ScCI (2). U2 (Cp*SiNR) (16.8 g, 64 mmol) and ScCI3 (THF) 3 (23.5 g, 64 mmol) were combined with 250 mL toluene and stirred overnight at room temperature. The resulting solution was filtered and the toluene was removed in vacuo, leaving an orange oil. Addition of petroleum ether (ca 100 mL) precipitated a white solid. Three crops were obtained. The solid was loaded into a flask and heated at 100*C under dynamic vacuum overnight to remove the THF. The product was then extracted away from the LiCI with hot toluene. The toluene was removed in vacuo, and petroleum ether was added to the resulting orange oil to precipitate 1 as a white powder (10.4 g, 51% yield). Elemental Anal. Found(Calcd) C, 54.14(54.45); H, 7.90(8.22); N, 4.23(3.93). IR(Nujol): 2733,2686,1465, 1372,1349,1320,1244,1198,1122,1076,1046,1029,849, 826,802,756, 692, 552, 500. (Cp*SINR) 8 cCH{ 8 l(CH 3 ) 3 )a (3). Compound 2 (3.0 g, 9.1 mmol) and UCH(TMS)2* 1/2Et20 (1.85g, 9.1 mmol) were combined in 150 mL toluene and stirred at 80*C for approx. ihr, then overnight at room temperature. The toluene was removed in vacuo and the residue was redissolved in petroleum ether and filtered to remove the LiCI. The product was then precipitated at -78*C as a white microcrystalline solid (3.12 g, yield 76%). Elemental Anal. Found(Caicd) C, 58.0(58.1); H, 10.0(10.2), N 3.0(3.1). IR(Nujol): 2756, 2663,1453,1372, 1360,1302,1232,1186,1023,860, 837,814, 790. 767, 756, 721,663,599, 581, 535.488. 78 [(Cp*SINR)(PM a 3 )Sc] 2 Ci-H ) 2 (4). Compound 3 (1.04 g, 2.3 mmol) was partially dissolved in 30 mL of petroleum ether in a thick-walled glass vessel, and 0.23 mL PMea (2.3 mmol) was added by vacuum transfer. H2 (4 atm) was added to the vessel and the reaction mixture was stirred at room temperature. After 15 min the solid was completely dissolved, leaving a yellowish solution. Upon continued stirring, a white precipitate formed. The reaction was stirred overnight and 4 was collected by filtration as a white powdery precipitate (0.71 g, 83% yield). Elemental Anal. Found(Caicd) C, 58.6(58.0); H, 9.8(10.0); N, 3.2(3.8). IR(Nujol): 2704, 2670, 1459,1375,1352,1335,1328,1279,1240,1299,1226,1200,1152,1130,1031, 1014, 950, 936,837, 814, 795, 738, 671, 662, 623. [(Cp*SINR)(PM e 3 )Sc] 2 (p, n2, n2 -C 2 H 4 (5). Compund 4 (0.308g, 0.414mmol) was dissolved in 10 mL of toluene in a thick-walled glass vessel. Two equiv. of ethylene (148 Torr, 104.3 mL), one equiv. per scandium, were condensed into the reaction mixture at -196°C. The reaction was thawed at -80aC and then stirred at 0*C for 2hrs. The resulting yellow solution was warmed to RT and then transferred to a flask attached to a swivel frit assembly in a nitrogen-filled glovebox. The solution was concentrated to ca 5mL and cooled to -80°C for 3 hr to afford orange-yellow crystals of 5 (0.170g, 53%). Elemental Anal, (oxidant added, two sets of data reported): Found C, 61.53,63.64; H, 9.31,9.71; N 3.07,3.19 Calcd. C, 62.8; H, 9.8; N, 3.25. [(Cp*SINR)Sc] 2 (p*CH 2 CH 2 CH 3 ) 2 ( 6 ). Compound 4 (0.531 g, 0.787mmol) was dissolved in 20 mL of toluene In a thick walled glass vessel. Two equiv. of propene (256 Torr, 104.3 mL), one equiv. per scandium, were condensed into the reaction mixture at -196°C. After warming slowly to 25 *C, the reaction was stirred for 15 minutes. The solution was then transferred to a flask attached to a swivel frit assembly in an In fille d glovebox. The solvent was removed under reduced pressure and the resulting white solid was suspended in 5 mL petroleum ether, filtered and washed once with another portion of petroleum ether to afford a white powder (0.156 g, 32%). Elemental Anal, (oxidant added, two sets of data reported): 79 Found C, 64.87,63.13; H, 10.07,9.76; N, 4.56,4.40; Calcd. C, 64.0; H, 10.2; N, 4.1. IR(Nujol): 2720,1455,1375,1350,1320,1240,1190,1120,1035,1010, 840, 815, 790, 760, 740, 470. [(Cp*SiNR)Sc] 2 (/i-CH 2 CH 2 CH 2 CH 3 ) 2 (7). The same procedure was used as is described above for 6, except that 1-butene (192 Torr, 104.3 mL) was added to a toluene solution of 4 (0.4 g, 0.54 mmol). The resulting white solid was redissolved in 10 mL toluene and the volatiles were then removed in vacuo. This cycle was repeated three times to ensure complete removal of PMea from the product Yield 0.173 g, 37.6%. Elemental Anal. Found(Calcd) C, 64.91(64.9); H, 10.09(10.3), N 4.16(4.0). IR(Nujol): 2720,1455,1378,1355, 1320,1243,1226,1196,1126,1096,1079,1038,1014,944, 838,814, 797, 762, 744,668, 532,497, 480. (Cp*SiNR)(PMea)ScCH 2 CH(CH 3 )CH 2 CH 2 CH 3 (8). 2.2 equiv. of 2-methyl-1-pentene (251 Torr, 104.3 mL) were condensed into a solution of 3 (0.50 g, 1.3 mmol) in 15 mL toluene at •196*C . The reaction mixture was warmed to RT and stirred for 2 days. Removal ofthe solvent under reduced pressure afforded a golden brown oil, which was dissolved in 2-3 mL pentane and cooled to *78 *C to afford white crystals (0.298 g, 50.3%). Elemental Anal. Found(Calcd) C, 62.7(63.3); H, 10.4(10.8), N 2.8(3.1). IR(Nujol): 2708,1461,1372.1355, 1308,1284,1237,1196,1138,1126,1032,1014, 961,950, 891,838, 814, 785, 756, 738, 662, 532, 520,497,479. (Cp*SINR)(PM e 3 )ScCH(C 8 Hs)CH 2 CH 2 CH 2 C 6 Hs (9). To a solution of compound 4 (0.20 g, 0.54 mmol) in 15 mL toluene were added 125 pL (1.1 mmol) styrene by syringe under the N2 atmosphere of a glove box. The reaction was stirred at RT for 1 hour. The volatiles were removed from the resulting orange-yellow solution under reduced pressure, leaving a yellow solid that was redissolved in petroleum ether. Cooling the solution to -78 *C afforded a bright, canary-yellow, microcrystalline solid (2 crops: 0.253 g, 81%). Elemental Anal. Found(Calcd) C, 70.22(70.4); H, 8.96(9.2), N 2.2(2.4). IR(Nujol): 2750,2700.1584,1455, 80 1376,1352,1308,1258,1199.1093,1026, 838, 811, 800, 747, 697, 597, 565. 535, 523, 500, 479. (C5 H 4 CM« 3 >SiM«2 CI (10). (CsH4 CMe3 )U (21.7g, 0.169 moles) was dissolved in 300 mL THF. The solution was cooled to -78 *C and MezSiCfe was added via syringe against a flow of argon. The reaction was warmed to RT and stirred overnight. The THF was removed under reduced pressure and the product was dissolved in 100 mL petroleum ether and filtered to remove LiCI. Removal of the petroleum ether in vacuo afforded a clear, light yellow, non-volatile liquid (33.1 g, 91%). U 2 (tBuCpSiNR) (11). Compound 10 was added dropwise by pipet to a solution of Li(Me3 CNH) in 300 mL THF with stirring (slightly exothermic) in a nitrogen-filled glove box. The reaction was stirred overnight at room temperature. The THF was removed under reduced pressure, and the resulting yellow-orange liquid was dissolved in 250 mL petroleum ether and filtered away from the LiCI. The petroleum ether solution of the ligand was cooled to -78*C and 118 mL of a 1.6 M solution of nBuU in hexanes were syringed into the solution under an argon flow. The solution was warmed to slightly above RT, at which point the evolution of gas was observed. The yellow reaction solution became tan colored and viscous. After 5 hr the petroleum ether was removed in vacuo to afford a tan, glassy solid (26g, 106% theoretical yield). Impurities resulting from the crude method of work-up probably account for the excess yield. (tBuCpSINR)ScCt(THF) (12). ScCl3 (THF) 3 (20.0g. 0.054 moles) and 11 (14.4g, 0.054 moles) were combined in a flask and 250 mL toluene were added to the solids. The reaction mixture was stirred at RT overnight The resulting dark-brown solution was filtered to remove LiCI. Removal of toluene in vacuo afforded a sticky, brown residue from which a solid precipitated upon addition of petroleum ether. The solid was suspended in 100 mL petroleum ether and collected by filtration. Repeated washings with petroleum ether to remove the soluble, brown impurities afforded a white solid (15.68 g, 87%). The solid may 4 81 be purified by recrystallization from toluene. Elemental Anal. Found(Calcd) C, 57.3(56.8); H, 8.71(8.8), N 3.22(3.5). IR(Nujol): 2720,2661,1455,1373,1243,1202,1179,1073,1032, 1011, 938, 920, 859, 832, 808, 790, 762, 738, 685, 672, 532,491. (tBuCpSINR)ScCH 2 PMe 2 (13). A flask was charged with compound 1 2 (3.0 g, 7.5 mmol) and LiCH2 PMe2 (0.61 g, 7.5 mmol). Toluene (60 mL) was condensed onto the solids at •78 °C, and the reaction mixture was warmed to RT and stirred overnight to afford a white powder, which coprecipitated with the LiCI co-product (2.66 g, 95%). Elemental Anal. Found(Calcd) C, 51.8(52.5); H, 8.2(8.6), N 3.1 (3.4). (tBuCpSlNR)(PMe 3 )ScCI (14). A thick walled glass vessel was charged with compound 13 (6.9g, 0.017 moles). Toluene (100 mL) was condensed onto the solid at -78 *C. The mixture was then frozen at -196* C and 1 equiv. of HCI (272 Torr, 1.5 L) was condensed into the vessel. The reaction mixture was warmed to RT and stirred overnight The solution was filtered through a pad of celite to remove the fine LiCI precipitate and then dried in vacuo to a foam. Recrystallization from petroleum ether afforded and off-white, microcrystalline solid (4.33 g, 64%). Elemental Ana). Found(Calcd) C, 53.37(53.2); H, 8.83(8.9), N 3.48(3.4). (tBuCpSINR)ScCI (15). Heating 14 to 120* C for 5 hr under vacuum liberates PMe3 to afford 14 cleanly as a foam. Elemental Anal. Found(Calcd) C, 54.01(54.6); H, 8.23(8.2), N, 4.25(4.2). IR(Nujol): 2719,2872,1455,1372,1296,1243,1196,1173,1079,1014, 955, 938, 914,832,808, 785, 767,750, 685,673, 538,491. (lBuCp 8 INR)(PMe 3 ) 8 cCH 2 TM S (16). Compound 14 (1.0g, 2.5 mmol) and UCH 2 TMS (0.23 g, 2.5 mmol) were dissolved in 15 mL toluene, and the reaction was stirred for 1 hr at RT. The toluene was removed under reduced pressure, and the product was dissolved in 3-5 mL pentane. Cooling the solution to -78 *C afforded white crystals (0.82 g, 87.4%). Elemental Anal. Found(Calcd) C. 57.9(57.7); H, 10.0(10.3), N, 3.3(3.1). IR(Nujol): 2720,2672,1455, 82 1377,1364,1353,1305,1286,1239,1198,1176,1082,1053,1030,1018,1004, 955, 940, 921, 832, 764, 746, 720, 662, 598, 537,496. [(tBuCpSlNR)ScM«]x (17). Compound 14 (0.50g, 1.52 mmol) and MeLi (0.035 g. 1.54 mmol) were measured into a flask and 20 mL toluene were condensed onto the solids at -78’ C. The reaction was warmed to RT and stirred for 2 hrs. The solution was filtered to remove LiCI, and the toluene was removed under reduced pressure. The residue was dissolved in 10 mL pentene and cooled to -78 *C to afford a fluffy precipitate, which was isolated as an off-white solid (0.264 g, 56%). Elemental Anal. Found(Calcd) C, 62.06(62.1); H, 9.94(9.8); N, 3.98(4.5). IR(Nujol): 2719, 2672,1455,1372,1296,1243,1196,1173,1073, 1020, 955, 932, 914, 632,806, 785, 761, 744, 726,685,673, 532,491. (,BuCpSINR)(PMe 3 )ScCflH5 (18). Toluene, 50mL, was added to compound 13 (2.0g, 5.3 mmol) in a thick walled glass vessel. One equivalent of HCI gas (942 Torr, l04.3mL) was admitted to the vessel at -196*C and the reaction was stirred overnight at room temperature. The resulting solution was transferred to a flask containing 0.67g (S.Ommol) phenyl lithium in a nitrogen-filled glove box. The flask was attached to a frit assembly and the reaction was stirred overnight The toluene was removed under reduced pressure, and the product was dissolved in pentane and filtered to remove insoluble salts. Concentration of the solution and cooling to -80*C afforded 0.433g (20% yield) of a tan solid. The product is extremely soluble and the reaction yield has not been optimized. Toepler experiment to determine amount of ethane produced In formation of 5 Compound 4 (45.5mg, 0.0612mmol) was dissolved in 2.5mL toluene in a small glass vessel (5mL "oligomerizer"). 0.12mmol of ethylene (57 Torr, 40.1 mL) was admitted to the frozen reaction mixture at -196*C. The reaction mixture was thawed at -80*C, stirred at 0°C for one hour and then at 25*C for 15 min. The volatiles from the reaction, including solvent, were vacuum transferred to a larger volume (ca 25mL) thick walled vessel and were kept at 83 - 8 *C during the Toepler measurement. The amount of gas measured (43±1Torr, 25.2mL; 0.58±.02mmol) corresponds to 95.5% the theoretical amount of ethane expected. IR analysis of the gas confirmed its identity as ethane. X-ray Crystal Structure Determination of [(Cp*SINR)(PMe 3 )Sc]2 (/J-H) 2 (4) A crystal grown from a concentrated toluene-cfe solution of 4 in an NMR tube was placed in a capillary in a nitrogen atmosphere and mounted on the diffractometer. Unit-cell dimensions and an orientation matrix were obtained from the setting angles of 25 reflections with 19°<20<21 °. The compound crystallizes in the monoclinic system, in space group C2/c (#15), with a = 21 .238 (3 )A, b = 11.470(2)A, c = 22.253(3)A, £ = 113.16(1)*, volume = 4984.0(14)A3; z = 8 (4 dimers), and density = 1.26 g cm'3. Two complete data sets were collected; the data were corrected for a slight decay, but not for absorption (/irnw = 0 . 1 2 2 , so neglect of this could have led to maximum errors of 1.3% in F). Lorentz and polarization factors were applied and the data were merged and then put on an approximate absolute scale by Wilson’s method. The position of the scandium atom was found from a Patterson map, and the rest of the molecule was located In a subsequent Fourier map. A few cycles of isotropic refinement lowered R to 0 .2 , and a difference map then showed the toluene molecule, disordered at a center of symmetry. Anisotropic refinement of all the heavy atoms reduced R to 7%, but the goodness of fit was 2.55. Hydrogen atoms were introduced at positions found on the maps calculated in their planes or (for the deuteriums of the toluene molecule) at calculated positions. Their positions and isotropic thermal parameters were refined (but not the parameters of the toluene deuterium atoms, which were repositioned once near the end of the refinement and were assigned isotropic B’s 20% greater than those of the carbon atoms they are bonded to). The final refinement involved 383 parameters in a single matrix; the largest shift/esd in the last cycle was 1.38 for z of C 8 ; shifts for all other atoms, including the hydrogens (except those on C 8 ) were less than 0.2 esd. The hydrogen atoms at C 8 were poorly defined in difference maps, which probably leads to their poor 84 refinement. Further cycles of refinement did not change the situation significantly. The final difference map showed maximum excursions of ±0.24eA-3; the final R-index for those reflections with Fo2>3a(Fo2) was 0.034. The labeling scheme is given in Figure 1.8, a summary of the data collection information is given in Table IV, coordinates and Ueqs for non-hydrogen atoms are listed in Table V, parameters for the hydrogen atoms are listed in Table VI, anisotropic thermal parameters are in Table VII, and complete distances and angles are given in Table VIII. X-ray Crystal Structure Determination of [(Cp*SINR)(PMe 3 )Sc] 2 (/i, *r2, n2-C 2 H 4 ) (5) Orange-yellow crystals were grown from a toluene solution of 5 cooled at -60 SC. A roughly cubic crystal was cut with a razor blade from a large hexagonal prism and held inside a 0.5 mm capillary with some hydrocarbon grease. The crystal was centered on the diffractometer and 25 reflections with 18*<20<21 * were found and centered. Preliminary cell dimensions and the orientation matrix were obtained from their setting angles; final cell dimensions were calculated from the setting angles of 25 different reflections with 32*<20<4O*. The compound crystallizes in the monoclinic system in space group C2/c(#15), with a = 23.706(5), b = 11.342(2), c = 21.141 (3)A, 0 = 111.57(1)*, and volume = 5286.2(17)A3; z = 4. Two equivalent sets of data were collected and merged to give the final data set; the R-factor for merging was 0.055. The data were corrected for a slight (0.2%) decay, but no absorption correction was applied because /irmax was so small, and the grease obscured the edges of the crystal. The structure was solved by MULTAN and refined by full matrix least squares. A toluene solvent molecule was found early on, disordered about a center of symmetry, and its carbon atoms were included as constant contributions to the structure factors, but their parameters were not refined. Hydrogen atoms attached to the methyl groups were positioned at idealized positions based on difference maps calculated in their planes; of the 12 methyl groups, 11 showed clearly 3 hydrogen atoms in reasonable positions. The last, C8, had 6 possible locations, so two sets of half-hydrogens were 85 introduced there. All these were fixed in position, with arbitrary (7.0A2) thermal parameters. The two hydrogen atoms on the bridging ethylene molecule were located in a difference map, and their positional and thermal parameters were refined. The refinement converged quickly; the final R-index for reflections with F02>3a(F02) is 0.069. The final goodness of fit, 3.84, is rather high; this probably reflects problems in modeling the disordered toluene solvate, because the three highest peaks, and 6 of the top 9, in the final difference map are in this region (0.7,0.7 and 0.5 eA*3); the next highest peaks are less than O.SeA"3, two near the Cp* ring and one near the t-butyl group on nitrogen. All other peaks are less than O^eA*3. There are 4 negative peaks in the difference map with magnitude greater than 0.4eA3; all are in the solvent region, the largest, -O.SeA*3, near C21. Unfortunately, the peaks and holes do not suggest a different model for this region. The labeling scheme is given in Figure 1.9, a summary of the data collection information is given in Table IX, coordinates and U«qa for non-hydrogen atoms are listed in Table X, parameters for the hydrogen atoms are listed in Table XI, anisotropic thermal parameters are in Table XII, and complete distances and angles are given in Table XIII. X-ray Crystal Structure Determination of [(Cp*SINR)Sc]2 (M*CH2 CH2 CH3>2 (6) Crystals of 6 were grown from a benzene solution cooled to ca 5 *C and mounted in glass capillary tubes. Most were twinned, but one was found whose rotation photograph was acceptable. The capillary was large and the crystal was obscured by grease inside it, so we could not determine its size or shape precisely. The crystal was centered on the diffractometer and unit cell dimensions plus an orientation matrix were calculated from the setting angles of 25 reflections with 20»2O *. Final unit-cell dimensions were obtained later from 24 reflections with 31 *<20<35*. The crystal was monocllnlc, space group P2-|/n (#14), with a = 9.429(2)A, b = 21 .937(5)A, c = 9.826(2)A, 0 = 99.39(2)*, and volume = 2005.2(7)A3; Z = 2. Two equivalent sets of data were collected; there was no decay in the crystal as measured by the three check reflections that we monitored during data collection. The data 86 were corrected for background, Lorentz and polarization factors were applied, and the data were merged to give one set that was put on an approximately absolute scale by Wilson's method. The locations of the scandium and silicon atoms were found by MULTAN and the remainder of the atoms were located in a subsequent Fourier map. Full matrix least squares, first with isotropic and then with anisotropic thermal parameters for the non-hydrogen atoms, converged with an R-index of about 15%. A disorder in the methyl atoms of the t-butyl group was noted and an alternate orientation of those three carbon atoms was added, with the population fixed at 20% of the total. Hydrogen atoms were added, either at calculated positions or based on difference maps calculated in their planes, on all carbon atoms except C20, the bridging atom, and the minor component of the t-butyl group. The hydrogen atoms were given isotropic thermal parameters 20% greater than the equivalent isotropic thermal parameter of the bonded carbon atom; neither their positions nor their thermal parameters, norths parameters of the t-butyl component (whose isotropic thermal parameters were fixed at the average of the equivalent isotropic thermal parameters of the three major carbon atoms) were refined. Further refinement converged with an R-index of about 9% and a goodness of fit of 2.65. Most peaks in the difference map (maximum height ±0.62eA'3) were in the region of the bridging n-propyl group, but a chemically reasonable model for an alternate bridging group could not be found. The two hydrogen atoms on the bridging carbon atom C20 were located, included as fixed contributions along with the rest of the hydrogen atoms (repositioned) and the minor component of the t-butyt group, and the refinement was completed. The final R-index is 0.0854; for those reflections with F02>3<7(F02), R s 0.0579. The goodness of fit is 2.38. The final difference map had excursions of +0.48 and -0.45 eA*3. The largest peak is about equidistant from the three atoms of the bridging propyl group, still suggesting that the model is inadequate, but the small size of the residual peaks indicates that whatever disorder exists in this region is minor. 87 The labeling scheme is given in Figure 1.10, a summary of the data collection information is given in Table XIV, coordinates and Ueqs for non-hydrogen atoms are listed in Table XV, assigned parameters for the hydrogen atoms and disordered t-butyl carbons are listed in Table XVI, anisotropic displacement parameters are in Table XVII, and complete distances and angles are given in Table XVIII. X-ray Crystal Structure Determination of (Cp*SiNR)(PMea)ScCH 2 TMS (16) Crystals of 16 were grown from a petroleum ether solution cooled at -80 *C and mounted in thin-walled glass capillaries. Preliminary photos of one crytal showed that it was single, with monclinic symmetry. Unit cell dimensions and an orientation matrix were calculated from the setting angles of 25 reflections with 13*<20<16*. The compound crystallizes in space group P2-|/n, #14, with a = 11.181(4), b = 18.206(5), c = 15.250(6)A, fi = 108.25(3)*, and volume = 2948.2(18)A3; Z = 4. Two complete data sets were collected out to 29 = 40*; the data were corrected for a slight decay, Lorentz and polarization factors were applied and they were put on an approximately absolute scale by Wilson’s method. No correction for absorption was made because i*rmn is small, and the two data sets merged together well. The structure was solved by direct methods, phasing the reflections by hand; and the E map calculated with about 300 reflections revealed the Sc, P, and Si atoms. MULTAN had great difficulty with the data. Once the Sc, P and Si atoms were located, the remaining heavy atoms were found in Fourier maps and were refined by six cycles of least squares. Hydrogen atoms were then introduced at calculated positions (C-H = 0.95A, with isotropic thermal parameters 20% greater than those of the bonded carbon atom), assuming staggered geometries at the methyl groups, as constant contributions to the structure factors. They were repositioned once, just before the final cycles of full matrix refinement, which adjusted the positional and anisotropic thermal parameters of all the non-hydrogen atoms and a scale factor (244 parameters). The final R-index for 4307 reflections with Foz>0 is 0.1298, somewhat high, but for the 2101 reflections with F02>3<t(F02) it is 0.0533, a more 88 acceptable number. The value for all reflections is high because of the large number of weak reflections (59% of the total). The final difference map had maxima of ±0.65eA3, with no discernible pattern to them. The largest positive peak was near C20, a methyl carbon with large, apparent thermal motion. The labeling scheme is given in Figure 1.11, a summary of the data collection information is given in Table XIX, coordinates and Ueqs for non-hydrogen atoms are listed in Table XX, assigned parameters for the hydrogen atoms are listed in Table XXI, anisotropic displacement parameters are in Table XXII, and complete distances and angles are given in Table XXIII. Calculations were done with programs of the CRYM Crystallographic Computing System and ORTEP. Scattering factors and corrections for anomalous scattering were taken from a standard reference (International Tables for X-ray Crystallography, Vol. IV, p.71, p. 149; Birmingham, Kynoch Press, 1974). R = 2 |F 0-|F C| |/ZF o, for only Fo2^ , and goodness of fit = [Ew(Fo2-Fc2)2/(n-p)] where n is the number of data and p the number of parameters refined. The function minimized in least squares was E w ffF o ^ c 2)2, where w = l/a ^ F o 2). Variances of the individual reflections were assigned on the basis of counting statistics plus an additional term 0.014I2. Variances of the merged reflections were determined by standard propagation of error plus another additional term, 0.014<l>2 89 References 1. a) Thompson, M. E.; Bercaw, J. E. Pure Appl. Chem. 1 9 84,5 6 ,1. b) Thompson, M.E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C.; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. J. Am. Chem. Soc. 1 9 8 7 ,109,203. 2. a) McKenzie, T. C.; Sanner, R. D.; Bercaw, J. E. J. Organomet. Chem. 1 9 7 5 ,102,457. b) Bercaw, J. E. Advances in Chemistty Series 1 9 7 8 ,1 67,136. c) Gibson, V. C.; Bercaw, J. E.; Bruton, W. J.; Sanner, R. D. Organometallics, 1 9 8 6 ,5 ,976 3. a) Lappert, M. F.; Holton, J.; Ballard, 0. G. H.; Pearce, R.; Atwood, J. L; Hunter, W. E. J. Chem. Soc., Dalton Trans. 1979,45. b) Lappert, M. F.; Holton, J.; Ballard, 0 . G. H.; Pearce, R.; Atwood, J. L; Hunter, W. E. J. Am. Chem. Soc* Dalton Trans. 1979, 54. c) Lappert, M. F.; Singh, A.; Atwood, J. L; Hunter, W. E. J. Chem. Soc., Chem. Common. 1983,206. d) Atwood, J. L; Smith, K. 0 . J. Chem. Soc* Dalton Trans., 1973,2487. 4. Burger, B. J.; Thompson, M. E.; Cotter, W. 0.; Bercaw, J. E. J. Am. Chem. Soc. 1990, 1 1 2 ,1566. 5. Parkin, G.; Bunel, E.; Burger, B. J.; Trimmer, M. S.; van Asselt, A.; Bercaw, J. E. J. Mol. Catal. 1 9 8 7 ,4 1 ,21. 6. The ebulliometric molecular-weight analysis gave molecular weights of 583 g/mole and 559 g/mole on two separate occasions (theoretical MW for monomer is 454.8 g/mole). 7. Bunel, E. Ph.D. Thesis, California Institute of Technology, 1988. 8. Pauling, L The Nature of the Chemical Bond, 3rd ed.; Cornell Press: Ithaca, 1960; Chapter 7. 9. Ghotra, J. S.; Hursthouse, M. B.; Welch, A. J. J. Chem. Soc* Chem. Commun. 1973, 66. 10. a) Brauer, D. J.; BOrger, H.; Essig, E.; Geschwandtner, W. J. Organomet Chem. 1980,343. b) Planalp, R. P.; Andersen, R. A.; Zalkl, A. Organometallics, 1983,12,16. c) Planalp, R. P.; Andersen, R. A. Organometallics, 1983,2,1675. 11. Lappert, M. F.; Pedley, J. B.; Sharp, G. J.; Bradley, D. C. J. Chem. Soc* Dalton Trans. 1976,1737. 12. Harris, R. K. Nuclear Magnetic Resonance Spectroscopy, Pitman: Marshfield, 1983; p. 232. 13. Evans, W. J.; Meadows, J. H.; Wayda, A. L ; Hunter, W. G.; Atwood, J. L J. Am. Chem. Soc. 1 9 8 2 ,104,2008. 14. Jordan, R. F.; Bajgur, C. S.; Dasher, W. E. Organometallics, 1987,6(5), 1041. 15. For other examples of p, rj2, fj2-olofin-bridged dimers see: b) Cotton, F. A.; Kibula, P. A. Polyhedron, 198 7 ,6 ,6 4 5 . c) Evans, W. J.; Ulibarri, T. A.; 90 Ziller, J. W. J. Am. Chem. Soc. 1 9 9 0 ,112,219. d) Bums, C. J.; Andersen, R. A. J. Am. Chem. Soc. 1 9 8 7 ,109,915. e) Ref. 17 16. Schultz, A. J.; Brown, R. K.; Williams, J. M.; Schrock, R. R. J. Am. Chem. Soc. 1981, 103 , 169 and references therein. 17. Kaminsky, W.; Vollmer, H. J.; Heins, E.; Sinn H. Makromol. Chem. 1 9 7 4 ,175,443. b) Heins, V. E.; Hinck, H.; Kaminsky, W.; Oppermann, G.; Raulinat, P. Makromol. Chem. 1 9 7 0 ,1 3 4 ,1. 18. Kaminsky, W.; Kopf, J.; Sinn, H.; Vollmer, H. J. Angew. Chem., Int. Ed. Engl. 1976, 75(10), 629. 19. Holton, J.; Lappert, M. F.; Ballard, D. G. H.; Pearce, R.; Atwood, J. L; Hunter, W. E. J. Chem. Soc., Dalton Trans. 1979,54. 20. Anderson, G. A.; Almenningen, A.; Forgaard, F. R.; Haaland, A. J. Chem. Soc., Chem. Common. 1971,480. 21. Huffman, J. C.; Streib, W. E. J. Chem. Soc* Chem. Commun. 1971,911. 22. Burger, B. J. Ph.D. Thesis, California Institute of Technology, 1987. 23. Doherty, N. M.; Bercaw, J. E. J. Am. Chem. Soc., 1985,107,2670. 24. In the reaction of triisobutylaluminum with styrene to form tris-(phenethyl)aluminum, 22-44% of the 2-phenethyl substituents were found in the products. This may be compared to a similar reaction with 1-hexene, in which n-hexyl products prevail. Natta, G.; Pino, P.; Mazzanti, P.; Longi, P.; Bemardini, F. J. Am. Chem. Soc., 1959, 2561. 25. a) Ewen, J. A. J. Am. Chem. Soc. 1 9 8 4 ,1 06,6355. b) Kaminsky, W.; KOtper, X.; Brintzinger, H. H.; Wild. F. R. W. P. Angew. Chem. In t Ed. Engl. 1 9 8 5 ,2 4 ,6. c) Kaminsky, W.; Bark, A.; Spiehl, R.; Mdller-Lindenhof, N.; Niedoba, S. In Transition Metals and Organometallics as Catalysts for Olefin Polymerization/ Kaminsky, W., Sinn H., Eds.; Springer-Vedag: Beriln, 1988; p. 291. d) Ewen, J. A.; Haspeslagh, L; Elden, M. J.; Atwood, J. L ; Zhang, H.; Cheng, H. N. In Transition Metals and Organometallics as Catalysts for Olefin Polymerization/ Kaminsky, W., Sinn H., Eds.; Springer-Vedag: Berlin, 1988; p. 281. 26. Burger, B. J.; Bercaw, J. E. In New Developments in the Synthesis, Manipulation and Characterization of Organometallic Compounds Wayda, A. L , Darensbourg, M. Y. £ds„- ACS Symposium Series, 357 (1987). 27. Marvich, R. H.; Brintzinger, H. H. J. Am. Chem. Soc. 1 9 7 1 ,9 3 ,2 0 3 . 28. a) Sharp, P. R. Organometallics, 1 9 8 4 ,3 ,1217. b) Wolfsberger, W.; Schmidbauer, H. Synth. React. Inorg. MeHDrg. Chem. 1 9 7 4 ,4 , 149. 29. a) Cowley, A. H.; Kemp, R. A. Synth. React. Inorg. M et-O rg. Chem. 1981, 77(6), 591. b) Davidson, P. J.; Harris, D. H.; Lappert, M. F. J. Chem. Soc* Dalton, Trans. 1976, 2268. 91 30. Karsch, H. H.; Schmidbaur, H. Z. Naturforsch, B: Anorg. Chem., Org. Chem. 1977, 32, 762. 92 Chapter 2 Investigations of the a-Olefln Polymerization Activity and Mechanism for Monocyclopentadlenyl Scandium Amido Complexes 93 Introduction Identifying the active species involved in industrially important catalytic processes is usually difficult and often impossible. For instance, the Fischer-Tropsch process for hydrocarbon synthesis, 1 a the Haber process for the reduction of to ammonia 1 *1 and the ammoxidation of propene to acrylonitrile1® are examples of processes whose mechanisms remain uncertain because the conditions of the reactions preclude direct characterization of the active catalyst species. The Ziegler-Natta olefin polymerization process is no exception. As mentioned in the general introduction, the catalyst centers are ill-defined because of their multicomponent, heterogeneous composition and the fact that usually only a small percentage of the transition metal centers are active. Even in homogeneous Ziegler-Natta systems based on mixtures of group IV metallocenes and an alkyl aluminum or aluminoxane cocatalyst, the active species are unidentifiable because of the presence of complex equilibria in the former case3 and the overwhelming amount of aluminum cocatalyst ( 1 0 0 0 fold!) necessary in the latter case. 3 One solution to better understanding the mechanisms of such hard-to-get-at catalyst systems is to develop well-defined, soluble, unimolecular complexes that undergo discrete reactions pertinent to the catalysis and are easily characterized by structurally informative methods such as NMR spectroscopy and x-ray crystallography. In fact, the development and investigation of such model complexes has been one of the mainstays of synthetic and mechanistic organometallic chemistry. There are now several well-defined, early transition metal and lanthanide complexes which, by virtue of their olefin insertion activity, serve as models for Ziegler Natta catalyst systems. The structural simplicity of these model systems serves to facilitate their mechanistic investigation. The permethylscandocene system is especially conducive to mechanistic study since its derivatives are monomeric and all activity is constrained to occur in the "equatorial wedge" between the cyclopentadienyl rings, where the frontier molecular orbitals are located.4 In increasing the activity of the scandium by progressing from the bis- 94 pentamethylcyclopentadienyl ligand framework to the monocyclopentadienyl amido ligand framework we sacrifice some of this simplicity. As a result of the decreased steric demand of the monocylopentadienyl amido ligand system, the scandium complexes tend to oligomerize and require coordinating ligands for stabilization. Our isolated, structurally characterized complexes are no longer the actual catalysts but precatalysts for the polymerization. It behooved us to determine the nature of the active catalyst (i.e., monomer vs. dimer, phosphine-bound vs. phosphine-free) if we were to understand our systems any better than the complex catalyst mixtures we were trying to model. We therefore set out to determine through kinetic analysis the nature of the active catalyst species derived from these precatalyst complexes. The results of these experiments are described within. All evidence points to a monomeric, Lewis-base free, 12e- scandium alkyl as the active chain propagating species. Other aspects of the polymerization reactions such as the molecular weights and polydispersities of the polymer products and the regioselectivity and stereospecificity of the polymerization are also discussed. In essence, this chapter presents our efforts to characterize the a-olefin polymerization activity of the complexes described in Chapter 1 . Results and Discussion General Features of Olefin Polymerization by [(Cp*SiNR)(PMe3)Sc]2(M~H>2(4). The a-olefin polymerizing activity of the (Cp*SiNR)Sc system was initially discovered with the scandium hydride complex (4). Most of the features of the polymerization using this precatalyst are summarized in Scheme I. As can be seen, the turnover rate of this system is rather slow. As a result, although the polymerization of propene may be observed in an NMR tube, the relatively low solubility of propene in toluene at 2 5 °C causes the synthesis of potypropene at monomer pressures of <5 atm. to be rather sluggish and impractical. Only modest molecular weight polymers are obtained from the polymerizations of 1-pentene and 1-butene at 25 °C. The molecular weights listed in Scheme I. 95 96 Scheme I were measured by GPC relative to polystyrene. The 1 -butene and 1 -pentene polymers are colorless, sticky oils, and resonances for vinylidene chain ends are observed in their 1H NMR spectra, implicating 0-H elimination as a principal chain-transfer pathway. Cooling the reaction results in an increase in the molecular weight of the polymer product and a slight reduction in the polydispersity index. This trend is expected since lowering the temperature at which the polymerization is performed should favor the intermolecular insertion reaction relative to intramolecular 0-H elimination. The sluggishness of the polymerization reaction prohibits cooling the reaction below 0°C , thus keeping us from our goal of eliminating chain transfer completely (as in a "living" system). As mentioned above, resonances that correspond to gem-disubstituted olefin (vinylidene) chain ends are observed in the 1H NMR spectra of polymers produced at room temperature by 4, consistent with chain transfer by 0-H elimination. We have found that gem disubstituted olefins (cf. 2-methyl-1-pentene) react slowly (1 day) with 4 in a stoichiometric fashion. This indicates that chain transfer by 0 -H elimination in this catalyst system is only slowly reversed, and reinsertion of the chain end does not compete effectively with the much faster addition of a-olefin. Also, branching of the polymer caused by the combination of vinylidene chain ends does not occur. Chain transfer by >9-alkyt elimination would lead to vinylic chain ends, which are not observed in the 1H NMR spectrum of polymers produced at room temperature. Such vinylic chain ends would not be observed, however, if they react in the same way as other a-olefins and are combined to form branched polymer. This possibility deserves investigation. A pattern in the olefin region of the 1H NMR spectrum that is diagnostic of vinylic chain ends is observed for polypropene prepared at 80 *C with 4. Thus, 0-alkyl elimination is evident at least at elevated temperatures. The scandium catalyst is not stable at this temperature, however, forming a new species that crystallizes from the solution as it cools. The structure of this decomposition product could not be elucidated from either its 1H or 13C NMR 97 spectrum, although the activation of PM8 3 C-H bonds is apparent. Efforts to obtain a crystal structure have been unsuccessful so that the identity of this decomposition product is still unknown. Further investigation is necessary to establish the importance of >9-alkyl elimination as a chain-transfer mechanism in this system. A search for this type of reactivity in model scandium complexes (cf. (Cp*SiNR)(PMe3 )ScCH2 CH(CH 3 )CH 2 CH 2 CH 3 (8 )) could be informative. The reactivity of 4 toward other unsaturated substrates has been examined briefly. Conjugated dienes are at least oligomerized with this system, albeit slowly and only at elevated temperatures (80 °C).® As will be discussed, the phosphine-free systems, such as the scandium propyl dimer (6 ), are more efficient catalysts for polymerizing a-olefins and are better catalysts for the polymerization of conjugated dienes as well. The reactivity of 4 toward gem-disubstituted olefins has already been discussed in Chapter 1. Internal olefins, allene, terminal and internal alkynes react only stoichiometrically with 4, and the identities of the reaction products were not pursued. Analysis of Propene Oligomer Molecular Weight Distributions Two primary concerns associated with studying a catalyst system such as 4 are that a) only a minor component of the mixture is responsible for the catalysis, and b) more than one active species is present Clearly, either of these situations would present severe limitations to our ability to describe the system. Certain observations with regard to polymerization catalysis by 4 served to diminish these two concerns. That all of the scandium is involved in the polymerization is indicated by the disappearance of signals characteristic of 4 in the 1H NMR spectrum upon addition of olefin; these resonances are regenerated when the reaction mixture is hydrogenated. Also, the PDI's of the polymers produced by 4, as determined by GPC, are approximately 2, indicating that the catalyst mixture is homogeneous. 6 98 Nevertheless, a more reliable method was needed for demonstrating that our catalyst system was well behaved, i.e., that all of the scandium centers were acting alike. Therefore, as a more careful check, the molecular weight distributions of propene oligomers produced by 4 were examined at low monomer conversion, where chain transfer by 0-H elimination is essentially absent. Given that the rate of initiation is at least as fast as the rate of subsequent insertions and that all of the scandium centers are active and inserting monomer at a uniform rate (conditions for a "living" 7 system), the molecular weight distribution of oligomers produced by 4 should be adequately modeled by a Poisson function based on the ratio of monomer consumed to scandium present 8 The oligomerization reactions were carried out as described in the experimental section. The resulting scandium alkyls were cleaved with hfe and the molecular weight distributions of the saturated oligomers were analyzed by GC. Since, with the exception of the propene trimer, 2,4-dimethyihexane, authentic standards were not available for comparison with the oligomers, the identities of the oligomers out to C 2 7 were verified first by GC/MS analysis. In determining the relative amounts of oligomers in the GC traces, the response factors of the oligomers were assumed to scale linearly with their carbon number. In Figure 2.1, two representative oligomer distributions are compared with the calculated Poisson distributions on the basis ofthe amount of monomer consumed assuming that all scandium centers are active. Good agreement is observed between the experimental and the calculated distributions, indicating that most, if not all, of the scandium centers are participating in the polymerization and are propagating at a similar rate. An interesting sidelight to this experiment was our finding that the maximum theoretical number of separable head*to-tail (vida infra) stereoisomers could be observed by capillary GC for the shorter propene oligomers, two for the tetramer, 2,4,6-trimethylnonane, four for the pentamer, 2,4,6,8-tetramethylundecane, and eight for the hexamer, 2,4,6,8,10tetramethyltridecane (Figure 2.2). Apart from the physical characteristics of the polymers 99 CO w > it 8,1 ® E c s o co .5 s £ ? CM o 0 1 E CO <A s co 3 llsr 5 z g 8 ? I § 5 0 its 25 w 3 t . i ^ »ts® - <0 C A N 5 £ 5 «r ? c © « E ■= Ilf <0 lit 'C •= rj'® = l& •S C o 3 O £ o».2» t t C o o CO o co U O jP B JJ o 6 |0 U i 0 A U B |0J o 100 u u e m a 1.0E6 8) 5.0E 5- X JO O | A ___ HI £ 015 16 17 T 1« • 19 ( m1n . ) 5 .0E 6: 3 . 0 E6 2.0E 6 1.0E6 25 26 Time ( m i n . ) 2.0E 61.5E51.0E6 £ 5.0E5 ( atm. ) Figure 2.2. GC analysis of a mixture of propens oligomers prepared with [(Cp*SiNR)(PMe3 )ScH]2 . Expanded regions show peaks corresponding to the maximum number of separable stereoisomers for a) 2,4,6-trimethylnonane, b) 2,4,6,8-tetramethylundecane and c) 2,4,6,8,10-pentamethyltridecane. 101 produced by 4, this was the first evidence that atactic polymer is produced by this catalyst system. Results from the examination of polymer tacticity by 13C NMR are presented later in this chapter. Kinetic Analysis of 1 -Pentene Polymerization by 4 Confident that 4 was indeed the catalyst precursor and that the system was well behaved, we proceeded to examine the kinetics of the polymerization in order to elucidate the nature of the active chain carrying scandium species. Since the concentration of catalyst centers is constant during the polymerization, assuming that the rate of olefin insertion is independent of the size of the alkyl chain on the scandium, the disappearance of olefin monomer can be monitored to determine a value for koba in the rate expression for the reaction (Equations 1 and 2). A series of kinetic experiments were carried out in this manner by monitoring the rate of disappearance of 1-pentene by 1H NMR. The concentration of scandium and the concentration of PMe3 were varied separately to determine their effect on kobsi thereby revealing whether the active species was monomeric or dimeric and whether phosphine loss was a prerequisite for olefin insertion. (1) [ScR] + CH2= C H R * — Si— [ScCHaCHRRl + CH2= C H R ’ -------— ► etc. (2) Rate - k,[ScRl fCH2^ C H R ’| - kob.(CH2= C H R ,l We began by varying the concentration of 4 in the reaction to determine its effect on kobt. As shown by the plots in Figure 2.3, the reaction was monitored for six different concentrations of 4, ranging from 0.017M to 0.072M in scandium. Clean first-order disappearance of monomer was observed over 3-4 half lives, confirming the assumption that the rate of propagation is independent of the size of the alkyi chain on the metal. The k0bs corresponding to each concentration of 4 is obtained from the slope of each line. When these values for k0 bSare plotted against [ScH] (Figure 2.4), instead of increasing linearly f [1 -pentene] Ul 1 102 0.017M j 0.024M [SC] - 0.0 I48M 0.038M 0.06M 10000 0 20000 30000 40000 50000 60000 o.oe 0.07 0.08 time (sec) Figure 23 kobsx exp-4 (1 /s e c ) 1.4 1.0 0.S 0.4 • 0.2 02 0.03 0.04 o.os [(Cp*SiNR)(PMCs)ScH) (M) Figure 2.4 103 with the concentration of scandium, they level off gradually. This behavior signifies a dissociative pre-equilibrium in the generation of the active species. The proposed mechanistic model for the polymerization (shown in Figure 2.5) involves a dissociative pre-equilibrium to afford the active catalyst (Cp*SiNR)ScR. This model is consistent with behavior observed in other Ziegler-Natta model systems. For example, kinetic studies of the insertion of propene by Cp* 2 LuCH3 * L revealed a dissociative pre-equilibrium to afford the active species Cp* 2 LuCH3 .® The monomeric nature of the active species was confirmed by kinetic analysis of the reaction of [Cp*2 l-uCH3 ] 2 with propene. Likewise, Jordan has shown that the active species in ethylene polymerization by [Cp2 Zr(CH3)(THF)+] is the "naked alkyl", [Cp2 Zr(CH3)+].10 [LScR] - [(Cp'SiNRKPMeaJScR] K, [LScRJ [ScRJ ♦ L k [ScR] ♦ CH2= C H R -------— - (ScCHjCHRRl Kl [LScCHjCHRRl _ = * [ScCHjCHRRl ♦ L (ScRJ - [(Cp*SiNR)ScRJ L-[PM#J [ScCHjCHRRl ♦ CHja-CHR* — - — - [ScCHaCHRCHjCHRR] etc Figure 2J. Mechanism of Olefin Polymerization by (Cp*SlNRXPMej)ScR We were already aware of the inhibitory effect of phosphine on the rate of olefin polymerization in that adding excess phosphine to the catalyst system slows the polymerization and the phosphine-free Sc-alkyl dimers are much more active than 4. To check the validity of the above model and to determine values for «i and ki, the kinetics of 1pentene polymerization by 4 was measured at 32" C for different concentrations of added PMe3 , with the concentration of scandium held constant Once again, clean first-order 104 f [1-pentene] w 1 disappearance of olefin was observed (Figure 2.6). o c ® 0.091 M & 0.07M J 0.049M l - o.oo; 0.0 2 8 M 100000 200000 time (sec) Figure 2.6 If we assume that the amount of dissociated phosphine is negligible relative to added phosphine. we obtain the expression for fc| shown in Equation 3. Given this assumption, a plot of 1/koba vs. [PMea added] should afford a straight line (Figure 2.7). The second-order rate constant, ki, is evaluated from the intercept of the line as 0.0015M'1sec'1. Using this value for ki, a value for Ki of 0.0085M is determined from the slope of the line. While an inverse dependence of koba on phosphine concentration is clearly indicated from the plot of 1/koba vs. [PMe3 added], the degree of phosphine dissociation indicated by the size of Ki is higher than initially estimated, too high for dissociated phosphine to be neglected in our calculations.11 Including dissociated phosphine into the equilibrium expression gives a more complex dependence of k0ba on added phosphine (Equations 6 and 7). This time, a non- 105 0.8 0.6 slope » 1 1 . 2 x 1 0 5 w ’ ssc - 1/fciK,[ScIo 0.2 irtwoopt - 9.7S X 10* S80*1 • i/ki[Sc|» 0.0 o.oo 0.08 0.04 0.02 [PMO) added] (M) Figure 2.7 (3) (4) (5 ) [(Cp*SiNR)ScR] [ L] K, - -------------------------------[Sc]e - [(Cp*SINR)ScRJ kob, - k1[(Cp*SiNR)ScR] 4 - K>t» - T 75Z T k,[Sc]0 + kiK^ScJo K, + [L] L k^JS e), [L] - [PMe3 added] * ltotal scandium] 0.08 106 linear, least-squares fit of the data gives a smaller value for K of 0.0012(5)M along with a second-order rate constant ki of 0.0094(40) M'1sec'1.12 A plot of the data (k0bs vs. [PMe3 added]) with the corresponding curve fit is shown in Figure 2.8. [ScR] ( [ScRl ♦ [L ]) K i a ------------------------------ (6) [ScJ0 - [ScR] kob* - M S c R ] » (k^/2) ( -([L] + + { ( [L] + K ^ 2 + 4K 1ISc]0}°-5) (7 ) Although an excellent fit with the kinetic data is obtained with this model, it was of concern to us that the value for ki determined at 3 2 *C from these experiments is essentially the same size as the value for ki (0.016M*1sec'1) determined for 1-pentene polymerization at -6 °C by the phosphine-free scandium propyl complex (6) (vide infra). Theoretically, if the same active intermediate is involved, the value for ki at 32 “C, using precatalyst 4, should be about a factor of ten greater than ki at -6*C , using precatalyst 6 .12 W e considered the possibility that our active catalyst was actually the phosphine-coordinated scandium species; however, the behavior of k ^ t expected from this model is completely different from what is observed. Yet another refinement was added to our mechanistic model in order for it to be complete and in order to improve our estimates of the parameters Kf and fc|. We included in our model the monomer/dimer equilibrium for (Cp*SiNR)ScR (Equation 8), which was revealed in the kinetic analysis of 1-pentene polymerization by 6 (vide infra). 107 O ■o P 2 <n Q. O ‘O £ 0) •O c *o o < s § oo CM O <d oo m CD •M i L_oas sqo>i CM o 108 k2 [ (ScR ) 2 ] 2 [ScR] (8) (ScR )2 (Cp*SiNR)Sc Sc(Cp'SiNR) A cubic equation (Equation 9) was derived, which involves a third parameter. Kg, to describe the new dependence of k0bs on added phosphine. [Sc] 0 - [ScR] + [(ScR)2 ] + [ScRL] [ScR]( [Sc] 0 + [L ]) * --------------------------(K2 + [ScR]) 2[ScR] + Kt + [ScR] (9) m t^i^obs]( [Sc] 0 + [L ]) ^ 2[ki/kobs] 2 (K2 + [ki/KobJ) K, + [^ /k o b J Given the limited set of data, a non-linear, least-squares fit to all three parameters was not feasible; therefore, a reasonable value for K2 at 32*C was calculated and included in the least squares analysis to estimate ki and Ki. A value for K2 at 3 2 *C of 0.039M was extrapolated from the value of 0.006M measured at -6* C by assuming a AS for dimer dissociation of 20 eu. Inclusion of the monomer/dimer equilibrium in the analysis serves to increase ki to 0.018(15)M*1sec'1 and to reduce Ki to 0.0006(5)M. The data and new curve fit are shown in Figure 2.9. Polym erization Rate Dependence on Added PMe$ + I o 0 .9 K1 = 0.0006M 0.8 ki s 0.018 M'1sec' 0.6 0 .5 «• a o 0 .4 0 .5 0.2 0.0 0.01 0 .0 2 0 .0 3 0 .0 4 0 .0 5 0 .0 6 0 .0 7 0 .0 8 0 .0 9 0 .1 0 Cone. Added PM ej (M ) Polym erization Rato Oopondenco on Added PM ej i o 1.6 ^ s 0.0004M 1.4 ki a 0.029 M^m c 1 1.2 o 4> CO ! \ 1.0 \ \ 0.8 ’ 0.6 \ • \ \ \ 0 .4 • 0.2 0.0 0.01 ngurafclO 0 .0 2 0 .0 3 0 .0 4 0 .0 5 0 .0 6 0 .0 7 Conc- Added P M * 3 (M ) 0 .0 8 0 .0 9 0.10 110 Despite our refinement of the model, the value for kt had not increased significantly. When another point is added to the data set, (the rate of 1-pentene polymerization by system 4 in the absence of added phosphine determined in a separate experiment), the value for ki determined from the non-linear, least-squares fit increases to 0.029(13)M‘1sec'1 (Figure 2.10). While this value is fairly stable to changes in the value for scandium concentration, it is quite sensitive to changes in the concentrations of added phosphine. Changing the value for.f<2 has a noticeable effect on ki as well.14 Although we were not able to resolve the discrepancy in the values for ki determined at the two temperatures, there are a number of assumptions and numerous sources of error in this experiment that could contribute to the analysis of ki. Nevertheless, a pre-equilibrium for phosphine dissociation is clearly Indicated. The same monomeric, phosphine-free scandium-alkyl species responsible for the polymerization by 6 (below) is, most likely, also the active intermediate in the polymerization by precatalyst 4. General Features of the Olefin Polymerization by [(Cp*SINR)Sc] 2 (/*-alkyl) 2 (6 and 7) Not only is an enhanced rate of olefin polymerization observed when the phosphinefree, scandium alkyl dimers 6 and 7 are the precatalysts instead of 4, but higher molecular weight polymer is obtained as well (Figure 2.11). Significant polymerization activity is still observed at temperatures as low as -10*C with the scandium propyl system. While lowering the temperature of the polymerization results in the formation of higher molecular weight polymer, we were disappointed to find that the molecular weight polydispersity of the polymer does not improve with the decrease in temperature. 111 [(Cp*SiNR)ScCH2CH2CH3J2 (Cp*SiNR)Sc polypentene: c a . 25°C; Mn - 6,000; PDI - 1.5 [neat pentene] polypropene: c a . 25°C; Mn - 5,200; PDI - 2.8 (ca. 25 v/v% in toluene) ca. -1 1°C; Mn - 12,700; PDI - 2.2 (ca. 25 v/v % in toluene) Figure 2.11 When the polymerization is monitored by 1H NMR, after the consumption of several equivalents of olefin relative to scandium, a considerable amount of the starting complex is unreacted. Thus, slow initiation relative to propagation could broaden the molecular weight distribution of the polymer. To confirm that it was indeed a leftover starting complex that we were observing in these polymerizations and not simply a generic spectrum for all phosphine-free scandium alkyls, a reaction mixture obtained from 6 and ten equivalents of propene was hydrogenolyzed and the volatiles were vacuum transferred to another NMR tube in order to measure the amount of propane liberated. Using hexamethyldisiloxane as a standard to quantify the propane by *H NMR, a 0.44 equivalent of propane relative to scandium was measured, indicating that 44% of the scandium propyl complex was unreacted after the ten equivalents of propene had been consumed. We attribute this slow rate of initiation relative to propagation to the stability of the alkyl bridges of the dimer complex, which must cleave to form the active 12e- scandlum-alkyl monomer. While following the kinetics of 1-pentene polymerization by 6, a uniform decay in the activity of the catalyst systems over time was observed. This catalyst decay would also tend 112 to broaden the molecular weight distribution of the polymer. The cause of this decay in activity has not yet been determined. Activation of the toluene solvent by the scandium was ruled out as the cause, since similar behavior was observed in methylcyclohexane and in cyclohexane. No olefin resonances that were due to /J-H elimination were apparent in the 1H NMR of polypropene produced by 6. Either the olefin chain ends are too dilute relative to the size of the polymer to be observed, or alternate forms of chain transfer and chain termination are more important in this system. Unfortunately, even the phosphine-free (Cp*SiNR)ScR complexes are not potential “living" catalysts for the polymerization of a-olefins. Nevertheless, the enhanced activity of these systems has allowed the preparation of polypropene for microstructure analysis (vide infra) as well as the polymerization of substrates, such as conjugated dienes,* which are only poorly polymerized by 4. As will be discussed in the next section, and has been alluded to earlier, these complexes have also allowed us to complete our mechanistic picture of the polymerization by enabling the dependence of the rate on the concentration of scandium to be measured in the absence of phosphine. Kinetic Analysis of 1-Psntsns Polymerization by 6 The kinetic analysis of 1-pentene polymerization by 4 established an inverse dependence of the rate on PMea concentration. Because of the complexity of the system, however, meaningful information could not be extracted from these data about the dependence of the rate on scandium concentration in order to determine the molecularity of the catalyst. The PMe3 -free scandium-alkyl dimers provided us with this opportunity. The problem of slow initiation by these systems had to be circumvented in order to ensure that activity from all of the scandium centers was being measured. Bulkier olefins such as neohexene and 3-methyM-butene were investigated as substrates which, by slowing the 113 polymerization, could narrow the gap between the rate of initiation and the rate of propagation in addition to making the reaction slower and easier to monitor. Interestingly, neohexene was unreactive toward these systems. 3-Methyl-1-pentene reacts with 6 and 7 but is not polymerized, so its use as a monomer was not pursued. Our ultimate approach was to use highly dilute catalyst solutions (ca 0.03M to 0.07M in scandium dimer) to favor scandium-monomer formation and a large excess of olefin (100 to 400 equivalents), so there would be plenty of olefin left over after all of the starting dimer complex had reacted. Upon monitoring the reaction at 32 *C , instead of the usual first-order disappearance in monomer, we found that the rate of the polymerization decayed over time. The source of this decay in catalyst activity was not determined, although it was slowed sufficiently at -6*C so that polymerization rates could be obtained. The first order disappearance in olefin over 2-3 half-lives is plotted for three different concentrations of scandium in Figure 2.12. A less than first-order dependence of koba on scandium concentration is evident from these data as well as from the plot of koba vs. [Sc-dimer] (Figure 2.13). In fact, the observed behavior corresponds to a monomer-dimer pre-equilibrium prior to olefin insertion modeled by Equation 10. This establishes the 12e-, monomeric scandium alkyl as the active species in the polymerization reaction. The non-linear, least-squares fit to the data gave values for the parameters ki and Kg of 0.016W1sec'1 and 0.008M respectively. A Brief Look at Olefin Polymerization by (tBuCpSINR)ScR Because of the recent development of the (,BuCpSINR)Sc system, our examination of its polymerization activity has been only preliminary. As mentioned in Chapterl, only the PMeyfree species are active, and [(tBuCpSiNR)ScMe]x is one complex we have managed to isolate for examination. Qualitatively, this system was found to polymerize olefins as slowly as the catalyst system based on 4. The dissociation of the scandium methyl oligomers to monomeric species is probably rate determining. Nevertheless, we were able to prepare polypropene with this system for stereochemical analysis (vida infra). GPC analysis of the 114 O" 1[euetuecH] l" 1[euejued-i] 115 o Q 10 in O r» X 0.03 0.02 0.01 0.00 k2 [(ScR)d 2 [ScR] [ScR] + Ic H 2 = C H R '| [(ScCHaCHRROaJ ----------- — [ScCH2CHRR1 Ka — “ 2[ScCH 2CHRR1 [SoCHjCHRRI + |CH 2 = C H R , | ----- - — - [ScCH2 CHR'CH2CHRR1 etc: Rate - It,(ScR] [c h 2 = C H R ’ 1 Kjb« ■ It, [ScR] [Sclo - ( ^ k , 2) ^ 2 + d/KOdtobJ (10) 116 polypropene produced at room temperature over two days by this catalyst revealed a fairly low number-average molecular weight of 3000 and a PDI of 2.1 S. Regio- and Stereochemistry of Olefin Insertion High regio* and stereospecificity are distinguishing features of Ziegler-Natta catalyst systems that have allowed the synthesis of commercially important, crystalline poly-aolefins.1® Since the regio- and sterocontrol of the polymerization is intimately linked with the mechanism of olefin insertion by Ziegler-Natta systems, current mechanistic models for both heterogeneous and homogeneous systems have relied heavily on the analysis of polymer microstructure.1® As a result, significant advances have been made in the elucidation of polymer microstructure, especially through 1H and 13C NMR spectroscopy.17 Described below are our efforts to characterize the regiochemistry and the stereochemistry of a-olefin polymerization by our monocyclopentadienyl scandium amido systems. The results obtained can be rationalized within the context of the behavior observed with other Ziegler Natta catalyst systems, and they contribute to our understanding of how ligand sterlc environments can influence selectivity. Regloselectivtty It is difficult to sort out the importance of electronic vs. sterlc effects in deciding the origin of regioselectivity in Ziegler-Natta systems. As a result, this aspect of the polymerization reaction has been somewhat taken for granted. In general, most heterogeneous and homogeneous Ziegler-Natta systems, including the group IV metallocene systems, are highly head-to-tail regioselective.18 It is not surprising, therefore, that our organoscandium model systems are highly head-to-tail regioselective as well. In the catalytic a-olefin dimerization by the silylene-bridged scandocene systems, {(i75-C9Me4)2 SiMe2 }ScH(PMe3 ) and [{(>;s-CsH3 CMe3 )2 SiMe2 }ScH]2 , only the head-to-tail dimerization products are observed.18 Primary insertion is indicated by the position of the 117 double bond formed in the /9-hydrogen abstraction step. The regioselectivity of olefin insertion by these systems can be attributed to steric control from the cyclopentadienyl ligands as well as to the electronic preference for locating the partial positive charge on the substituted carbon in the transition state for olefin insertion into the scandium-alkyl. To examine the regioselectivity of olefin insertion by (Cp*SiNR)ScR systems, the propene trimers formed with [(Cp*SiNR)(PMe 3 )ScH ] 2 were examined by GC. These trimers were generated by reacting the catalyst with three equivalents of propene, cleaving the scandium alkyls with H 2 and collecting the volatiles. Of the four possible C 9 regioisomers (Figure 2.14), only the head-to-tail product, 2,4-dimethylheptane, was observed. This result translates into >99% head-to-tail regioselectivity in this system. That this regioselectivity results from primary, or "1 - 2 ",olefin insertion is indicated by the structure of the single­ insertion product [(Cp*SiNR)Sc] 2 (/i-propyl) 2 as well as by the appearance of vinylidene chain ends in the 1H NMR of the polymer. [(Cp*SiNR)(PMea)ScH]2 + 3 Figure 2.14. Regioselectivity of Propene Insertion by (Cp*SINR)ScR 110 The regioselectivity of a-olefin insertion by the (tBuCpSiNR)Sc system has not yet been investigated. While largely head-to-tail material is indicated in the 13C NMR spectrum of polypropene produced with [(tBuCpSiNR)ScMe]x (vide infra), the presence of broad, less intense peaks outside of the designated methyl, methylene and methyne regions suggests the presence of some regioirregular monomer units. 1 9 Reduction in regioselectivity is consistent with the decreased sterics of the tBuCpSiNR ligand environment as compared to the Cp*SiNR ligand environment. A GC analysis of propene trimers similar to that described above for the (Cp*SiNR)Sc system is the next step for determining the prevalence and nature of the misinsertions. Stereospecificity The stereochemistry of a-olefin insertion can be influenced by the chirality of the last inserted monomer unit in the growing chain (chain-end control), the chirality of the metal center (enantiomorphic site control), as well as by a combination of the two effects.15a The relative importance of the two mechanisms of stereocontrol can be determined by examining the effect of the occasional stereochemical defect on the subsequent stereochemistry of the polymer, by examining the stereochemistry of a-olefin insertion following a non-stereoactive ethylene unit and by examining for the stereoselective polymerization of racemic monomers. These methods demonstrated early on that isotactic specific polymerization in heterogeneous systems originates from enantiomorphic site control and regiospecific “1 - 2 " insertion, leading to models concerning the structure of the asymmetric, heterogeneous sites. Through the development of the chiral ansa-metallocene systems (ethylene-bistetrahydroindenytyMCb (M = Ti, Zr, Hf) by Brintzinger and coworkers (Figure 2.15),2 0 the homogeneous, isotactic specific polymerization of propene became possible. This was the first example of the deliberate control of polymer microstructure through ligand design. The C2 symmetry of the complex, which is responsible for this stereospecificity is reminiscent of 119 the C 2 structures proposed for heterogeneous, isotactic-specific sites of TiCl3 -based catalyst systems. Other substituted metallocene systems having C2 symmetry and varying degrees of isospecificity have since been reported.2 * isotactic polypropene Figure 2.15 Another metallocene system, isopropyl(cyclopentadienyM-fiuorenyl)hafnium(IV) dichloride (Figure 2.16), reported by Ewen ef a/ . , 2 2 offers the first example of syndiospecific propene polymerization by an enantiomorphic site-control mechanism. The selectivity arises from size disparity between the ring systems on opposite sides of the metal. In contrast to earlier syndiospecific systems, 6 which displayed chain end control through a secondary mode of insertion, this system inserts olefin in a "1-2” fashion. Not surprisingly, the symmetric, normal bis-cyclopentadienyl metal systems do not display site stereocontrol and produce atactic polymer.2 3 120 syndiotactic polypropene C 3 N M - Hf, Zr Figure 2.16 Our (Cp*SiNR)ScR system is analogous to the mixed-ring complex of Ewen's in that the active species is achiral, and two different ligands of presumably different steric demand (the tetramethylcyclopentadienyl ligand and the tert-butyl amido ligand) define the active site for olefin insertion. Therefore, we anticipated that some syndiospecificity would be observed with this system. To investigate this possibility we examined the 13C NMR spectrum of polypropene prepared with [(Cp*SiNR)Sc]2 (p-prop)2 < 13C NMR spectroscopy is a powerful method for the elucidation of polymer microstructure and has found extensive use in the stereochemical analysis of polypropene.10c_* Shown in Figure 2.17 is the methyl region (with pentad assignments) of a sample 13C NMR spectrum for somewhat isotactic-rich, head-to-tail polypropylene.20 * The methyl region erf the spectrum is more sensitive to differences in stereochemistry than the methylene and methyne regions of the spectrum, exhibiting a wider range of chemical shifts. From the relative intensities of the pentad bands, one can perform a statistical analysis or "mapping" of the polymer micrcstructure to evaluate the mechanism of stereocontrol in the catalyst system. These pentad bands, as shown, occur as fine structure to the larger regions corresponding to the mm, the mr and the rr triads. 121 mm 23 .0 3 2 .0 31.0 30.0 1 » .0 Figure 2.17. Sample 13C NMR spectrum of polyprepene showing methyl region with pentad assignments. Shown in Figure 2.18a is the methyl region of the 13C spectrum of polypropylene prepared at 25 °C with [(Cp*SiNR)Sc]2 (M'Prop)2 < While the spectrum reflects atactic stereochemistry, there is a noticeable preponderance of nr triads over mm triads indicating that the catalyst is, to a small degree, syndiospecific. Although the 13C data are only preliminary and not rigorously quantitative, the band intensities for the methyl carbons should, to a first approximation, represent the relative amounts of the different stereoisomeric sequences. The relative triad and pentad band intensities do not appear to fit neatly into either the Bemouilian or the enantiomorphic-site control models for polymer stereochemistry. Thus, some combination of sterocontrol from the metal complex as well as from the adjacent stereocenter of the growing chain is involved in determining the stereochemistry of the polymer. 122 1S.0 18.3 18.0 1 7 ." 17.0 PPM ir s Figure 2.18. a) Methyl region of 13C NMR spectrum of polypropene prepared at 25°C with [(Cp*SiNR)Sc(p-CH 2 CH 2 CH 3 )]2 . b) Methyl region of 13C NMR spectrum of polypropene prepared at -10°C with the same catalyst. 16.3 123 Since the effects of site control and chain end control should be opposite in this system, it was not immediately obvious how lowering the temperature of the polymerization would affect the stereochemistry of the polymer. As can be seen from the 13C NMR spectrum in Figure 2.18b, cooling the polymerization reaction to -10°C does not significantly change the stereochemistry of the polypropene formed. While our initial goal in preparing the chiral (,BuCpSiNR)ScR system was to promote the isospecific polymerization of a-olefins, it is clear from our results with (Cp*SiNR)ScR that the monocyclopentadienyl-amido ligand-framework does not exert a great deal of steric influence over the stereochemical course of the polymerization. Therefore, it was not surprising that the polypropene produced with [(tBuCpSiNR)ScMe]x is purely atactic, as can be seen from the essentially 1 :2 : 1 ratio of intensities for the mm, mr and rr triads in the methyl region of the 13C spectrum (Figure 2.19). The absence of stereocontrol from the tBuCpSiNR ligand system as well as the minimal stereocontrol observed with the Cp*SiNR ligand system can be understood by considering the x-ray crystal structure for (,BuCpSiNR)(PMe3 )ScCH2TMS discussed in Chapter 1 . The tetrahedral position occupied by the PMea marks the position that should be approached by the olefin in the transition state for insertion. This position is fairly removed from the steric influence of the cyclopentadienyl ring system as compared to the "equatorial wedge" of metallocene systems which is greatly affected by the sterics of the Cp rings. Hence, the syndiospecificity of (Cp*SiNR)ScR is not as pronounced as that observed with Ewen's mixed ring system, and the asymmetric environment provided by (tBuCpSiNR)ScR appears to have no influence whatsoever over the steric course of the polymerization. 124 ■ ■ ■ ■ ! ■ ■ at.o — — “ I —* ao.9 • ■ • 1 I a*.* I » " I 1 tt.s ^ « I * is.a ■ I I T ir s • ■ ■ ' I 1 t« * I ■ I I • tr.s I I ' I ta.o Figure 2.19. Methyl region of 13C NMR spectrum of polypropene prepared under ambient conditions with [(tBuCpSiNR)ScMe]x. It is clear from existing examples that through the rational design of the supporting ligand environment, one can control the microstructure of the polymer that is formed. Efforts in our group and in others are currently being directed toward understanding the detailed steric interactions responsible for the stereocontrol observed with these systems and perhaps developing formulas for ligand-controlled, stereospecific polymer synthesis. As part of our continuing program of ligand modification, we are seeking to develop a ligand system for scandium that has a greater influence over the steric course of the polymerization than we observe with the (Cp*SiNR)Sc and the (tBuCpSiNR)Sc systems. Work is already in progress in the development of a dimethylsilylene-bridged fluorenyi-amido ligand system, which is expected to have more stereodirecting influence in the formation of syndiotactic polymer. Summary and Conclualona Through the experiments described in this chapter we are able to make the following statements about a-olefin polymerization catalysis by our monocyclopentadienyl scandium 125 amido systems: 1) The catalysis originates from scandium centers formed directly from the scandium hydride or scandium alkyl precatalyst complexes, not from minor impurities in the mixtures. A homogenous mixture of active species is involved in the catalysis. In other words, all of the scandium centers are acting alike. While these conclusions are drawn from an oligomer molecular weight distribution study using [(Cp*SiNR)Sc]2 0 *-H)2 , they may be logically extended to the phosphine-free (Cp*SINR)Sc-R complexes and probably to the analogous [(tBuCpSiNR)ScMe]x system as well. 2) At least in the case of the [(Cp*SiNR)ScR] catalyst system, the active chain propagating species is the monomeric, 12e-, "naked" scandium alkyl. This was established through analysis of the polymerization rate dependence on PMe3 concentration, using 4 as the precatalyst and on scandium concentration using 6 as the precatalyst 3) The [(Cp*SiNR)ScR] system is >99% head-to-tail regioselective for the polymerization of propene. Lower regioselecitivity with [(t8 uCpSINR)ScMe]x is indicated in the 13C NMR of polypropene made with this system. 4) These dimethyisilylene bridged monocyciopentadienyi-amido ligand-frameworks do not exert much steric influence over the stereochemistry of the polymerization. For the most part, atactic polymer is produced, although a small degree of syndiospedfidty is observed with the (Cp*8 INR)ScR. The lack of stereocontrol is attributed to the remoteness of the ring system from the site of the incoming olefin. There are still a number of unresolved questions concerning the reactivity of these catalyst systems, especially with respect to the mechanisms of chain transfer that are operating and with respect to the cause of decay in the activity of the PMe3 -free [(Cp*SiNR)ScR ] 2 complexes. In addition, only preliminary observations concerning catalysis by the (t8 uCpSiNR)ScR system have been made. 126 Nevertheless, we can use the information we have obtained so far to improve our design of future a-olefin-polymerizing catalyst systems. For instance, a considerably bulkier ring system than tetramethylcyclopentadienyl and i-butylcyclopentadienyl would prohibit tight bridging interactions between the scandium centers as well as have a greater influence over the stereochemistry of the polymerization. Our ultimate and rather far-reaching goal is to design single-component organoscandium systems that catalyze the "living" polymerization of a-olefins in a highly stereocontrolled fashion. 127 Experim ental Section General Considerations. All manipulations were performed by using glovebox and high* vacuum line techniques as described elsewhere.2 4 Solvents were dried by standard techniques and further purified by vacuum transfer from titanocene2 ® or sodium benzophenone. Toluene-cfe was dried over activated 4A molecular sieves and stored over titanocene. Argon and hydrogen were purified by passage over MnO on vermiculite and activated 4A molecular sieves. Propene (Matheson) was freeze-pump-thawed at least twice prior to use. PMe3 (Aldrich) was used without further purification. 1-Pentene (Aldrich) was dried over either titanocene or sodium sand. 1,2,4-Trichlorobenzene (Aldrich) was used as received. Cp2 Fe, used as an internal standard for NMR tube kinetic experiments, was sublimed before use. NMR spectra were recorded on Varian EM-390 (1 H, 90MHz), JEOL FX90Q(1H, 89.56), JEOL GX400Q ( 1 H, 399.78 MHz), and Bruker WM500 ( 1 3 C, 125.77MHz) spectrometers. GC data were measured on a Perkin-Elmer 8410 gas chromatograph using a 1 0 m x 0.53mm OV- 1 macrobore column from Alltech. GCMS data were obtained on a Hewlett Packard 5890A gas chromatograph interfaced with a Hewlett Packard 5970 series mass spectrometer, using a Hewlett Packard 12.5m OV- 1 capillary column. GPC analyses were performed on a Waters 150*0 ALC/GPC with five styragel columns (1 0 * - 1 0 0 A porosity). Non-linear, least-squares analysis of the kinetic data from the rate dependence of 1-pentene polymerization on added PMe3 was performed using the program Gauss, System Version 2.0 (Copyright 1988, Aptech Systems Inc., Kent, Washington) on an IBM pc. All other linear and non-linear, least-squares fits were performed with the program Cricketgraph on a Macintosh computer. 128 Sample Polymer Syntheses a)Poly-1-butene Synthesis by 4 To a solution of compound 4 (0.03g, 0.04mmol) in SmL toluene in a thick-walled glass vessel were added ca 1600 equivalents of 1-butene (790 Torr, 1.SL) at -196°C . After stirring for 7 days at room temperature, the reaction was quenched by the addition of 0.3mL methanol against an argon counterflow. The scandium was removed from the polymer product by extraction of the toluene solution with dilute HCI. The toluene phase was dried over MgS0 4 , and the toluene was then removed under reduced pressure to afford 2.5g of polymer with a number average molecular weight of 4000 and a PDI of 1.7 (GPC vs. polystyrene). b)Poly-1-pentene Synthesis by 4 Compound 4 (0.031 g, 0.042mmol) was dissolved in 10.8mL 1-pentene and the reaction was stirred at room temperature for 19hrs. The reaction was quenched with 2 mL methanol, and the excess pentene was removed under vacuum. The residue was dissolved in ethyl ether, extracted with 6 M HCI followed by concentrated NaHCQj and then water. The ether phase was dried over MgS0 4 , and the ether was removed under reduced pressure to afford Sg polymer with a number-average molecular weight of 2900 and a PDI of 2.08 (GPC vs. polystyrene). The reaction was carried out in a similar fashion at 0* in a constant temperature bath (37mg 4 , 10.2mL pentene) to afford after ca 25hrs. 0.044g polypentene (Mn = 7,000, PDI = 1.7). c)Polypropene Synthesis by 0 Approximately 0.016 mole propene (760 Torr, 0.4L) was was stirred over titanocene at -80*C and then condensed onto a solution of 6 (21.5mg, 0.032mmol) in 2.5mL toluene at -196*C . The reaction was stirred for approximately 8 hrs while submerged in a constant 129 temperature bath set at 24 *C. The reaction was quenched at this temperature by the addition of ca 1mL MeOH against an argon counterflow. The resulting toluene solution was extracted with dilute HCI, dried over MgS0 4 . and the toluene was removed under reduced pressure to afford 0.5g polymer. Mn = 5,206, PDI = 2.75 (GPC vs. polystyrene). The reaction was carried out in a similar fashion at a constant temperature o f-9 *C to afford 0.6g polymer (M„ = 16,500, PDI = 1.7). c)Polypropene Synthesis by [(tBuCpSINR)SeMe]x (17) Approximately 0.016 mole propene (760 Torr, 0.4L) was stirred over titanocene at -80 *C and then condensed onto a solution of 17 (19.8mg, 0.064mmol) in 2.5mL toluene at -196*C . The reaction was stirred for approximately 2 days at room temperature. The reaction was quenched by the addition of ca 1mL EtOH against an argon counterflow. A worir-up of the polymer similar to that described for the polypropene syntheses above afforded 0.2g polymer. Mn = 3060, PDI = 2.15 (GPC vs. polystyrene). Oligom er M olecular W eight Distributions. In a standard oligomerization run, propene (3-8 equiv.) was measured into a calibrated gas volume and condensed at -196*C into a small reaction vessel containing a 0.5 mL solution of 3 (ca 0.1 M) in toluene-da. The reaction mixture was stirred at room temperature for 1-2 hrs. The volatiles were then vacuumtransferred into an NMR tube at -196*C along with a measured amount of hexamethyidisiloxane. The amount of unreacted propene was determined by integrating it relative to the hexamethyidisiloxane standard at -80 *C by 1H NMR spectroscopy. The reaction residue containing the propene insertion products was redissolved in ca 1 mL benzene and stirred under 4atm Hz at room temperature overnight th e resulting mixture was exposed to air and immediately passed through a plug of silica-gel to remove all scandium byproducts, end the solution of saturated oligomers was analyzed by gc and gc/ms. Because of the unavailability of authentic hydrocarbon standards, gc/ms was used to verify the identity of the oligomers. The response factors of the oligomers, using flame 130 ionization detection, were assumed to be linearly related to their carbon number and were scaled accordingly. Kinetic Analysis of 1 -Pentene Polym erization by 4; Rate Dependence on Scandium Concentration In a nitrogen-filled glovebox, a stock solution was prepared by dissolving 100.3mg of 4 and 126.8mg of CpaFe in 3mL toluene-de. Serial dilutions of the stock solution were prepared in live different NMR tubes by adding the appropriate amount of toluene-de to a 0.2mL, 0.3mL, 0.4mL, 0.5mL and a 0.6mL aliquot of the stock solution to bring the final volume of each sample to 0.6m L Each NMR tube had a 14/20 ground-glass joint, which could be attached to a needle valve. To add 1-pentene to each sample, the NMR tube/needle valve assembly was attached to a 104.3mL gas volume on the vacuum line. The sample was frozen at -196*C , the NMR tube was evacuated and 240 Torr of 1-pentene were admitted to the gas bulb and then condensed into the NMR tube. The pentene was estimated to contribute an additional 0.15mL to the volume of the sample in calculating concentrations of catalyst and olefin. The tube was then sealed with a torch and stored at •196*C prior to the experim ent Each NMR tube sample was fully submerged in a 3 2 *C oil bath and removed at regular intervals to monitor the change in 1 -pentene concentration relative to the internal ferrocene standard on an EM 3901H NMR spectrometer, probe temperature 32 *C . Spectra were generally recorded over the duration of 3-4 half lives for each sample. K inetic Analysis off 1 -Pentene Polym erization by 4; Rate D ependence on Added Phosphine In a nitrogen-filled glove box, a stock solution was prepared by dissolving 133.9mg of 4 and 164.4mg of CpzFe in 4mL toluene-de. Five NMR tube samples were prepared by adding a 0.6mL aliquot of the stock solution to each tube. Each tube had a 14/20 ground- 131 glass joint, which could be attached to a needle valve. PMe3 was added to each sample by attaching the NMR tube/needle assembly to a 2.8(±0.4)mL gas volume on the vacuum line. The sample was frozen at >196*0, the NMR tube was evacuated and a diffent amount of PMe3 (30 Torr, 120 Torr, 210 Ton1, 300 Torr and 390 Torr) was admitted to the gas volume for each sample and then condensed into the NMR tube. 1-Pentene was then added to each sample by replacing the 2.8mL gas volume with the large 104.3mL gas volume, admitting 240 Torr of 1-pentene into the gas volume and condensing it into the NMR tube. Again, the contribution of added 1 -pentene to the total volume of the sample was estimated to be 0 . 1 SmL The tube was then sealed with a torch and stored at -196*C until use. Each NMR tube sample was fully submerged in a 3 2 * C oil bath and the kinetics were monitored over 3-4 half lives for each sample by 1H NMR on an EM390 instrument Kinetic Analysis of 1-Pentene Polym erization by 6 ; Rate Dependence on Scandium Concentration In a nitrogen-filled glovebox, a stock solution was prepared by dissolving 12.3mg of 6 and 45.5mg of CpzFe in 0.9mL toluene-de* Three NMR tube samples containing different concentrations of catalyst were prepared by serial dilution of the stock solution. The first tube contained 0.3mL of the stock solution, the second tube contained 0.1 SmL of the stock solution diluted with 0.1 SmL toluene-de and the third tube contained 0.075mL of the stock solution diluted with 0.225mL toluene-de. Each tube was attached to a needle valve via a ground-glass joint and frozen immediately in liquid nitrogen to minimize catalyst decomposition. Each NMR tube/needle valve assembly was connected to a l04.3m L gas volume on the vacuum line. The NMR tube, while still frozen, was evacuated, and 240 Torr of 1-pentene were admitted to the gas bulb and and then condensed into the sample. The tube was then sealed with a torch and stored at -196 * C prior to the experiment. The NMR probe of a JEOL FX90Q spectrometer was cooled to -6 *C , as calibrated by 132 a CH 3 OH standard. Prior to being loaded into the precooled probe, each sample was thawed in a -78* dry ice/acetone slush bath and shaken to redissolve solids that had precipitated from the solution. The concentration of 1-pentene in each sample was monitored relative to the ferrocene standard over two half-lives. The temperature of the probe was checked with a CH 3 OH standard before each sample as well as at the end of the experiment. Before and after the experiment the temperature of the probe was found to be -6 *C.. Before the second sample (containing 0.0075M scandium dimer) the temperature was measured as -2*C , and before the last sample (containing 0.0038M scandium dimer), the temperature was measured as -4*C . These apparent temperature fluctuations correspond to at most £ 3Hz change in the 166Hz chemical shift difference between the methanol peaks and are most likely due to errors in the chemical shift method of temperature measurement. Generation of Propane Trimers by 4 for GC Analysis Three equivalents of propene (23Torr, 132.6) were condensed from a gas bulb into a solution containing 20.0mg (0.054mmol) of 4 in 1-2mL benzene in a thick-walled vessel cooled to -196*C . The reaction was warmed to room temperature and stirred for one hour. H2 (4 atm.) was admitted to the reaction mixture, which was then stirred overnight at room temperature. The H2 was removed by freeze-pump-thawing the reaction, and the volatiles were vacuum transferred to a separate vessel and analyzed by GC. 13C NMR Analysis of Polypropene Samples Polypropene samples (0.2-0.5g) produced by [(Cp*SiNR)Sc]2 (/<-propyl) 2 and by [(tBuCpSINR)ScMe]x were dissolved in 2-3mL 1,2,4-trichlorobenzene and the solutions were placed in 10mm NMR tubes. A coaxial tube containing DMSO-de was used as an external NMR standard. The samples were heated to ca 80 *C in the NMR probe to obtain adequate resolution of the methyl pentad signals. Proton-decoupled 13C spectra were obtained with a pulse delay of 1 sec. Normally, 2000 scans gave sufficient signal to noise. 133 References 1. a) Sheldon, R. A. Chemicals from Synthesis Gas; 0 . Reidel: Oodrecht, 1983; Chapter 3. b) Ertl, G. Catal. Rev. 1 9 8 0 ,2 1 ,201. c) Gates, B. C.; Katzer, J. R.; Schuit, G. C. A. Chemistry of Catalytic Processes; McGraw-Hill: New York, 1979; Chapter 4. 2. a) Henrici-Oliv6 G.; Olivd, S. Angew. Chem., Int. Ed. Engl. 1 9 6 7 ,6 (9 ), 790. b) Breslow, 0 . S.; Newberg, N. R. J. Am. Chem. Soc. 1957, 7 9 ,5072. c) Breslow, 0. S.; Newberg, N. R. J. Am. Chem. Soc., 1 9 5 9 ,8 1 ,81. 3. a) Sinn, H.; Haminsky, W. Adv. Organomet. Chem. 1 9 8 0 ,18,99. b) Kaminsky, W.; Steiger, R. Polyhedron, 1988, 7(22/23), 2375. c) Glanetti, E.; Nicoletti, G. M.; Mazzocchi; R. J. Pofym. Sc/., Pciym. Chem. Ed., 1985,2 3,2117. 4. Lauher, J. W.; Hoffmann, R. J. Am. Chem. Soc. 1976,9 8,1729. 5. Cotter, W. D. unpublished results. 6. Doi, Y.; Keii, T. Synthesis of "Living" Polyolefins with Soluble Ziegler-Natta Catalysts and Applications to Block Copolymerization; Advances in Polymer Science; Springer-Verlag: West Berlin, 1986; Vol. 73/74; p.201. 7. For a discussion of living polymerizations, see: Swarc, M. Living Polymers and Mechanisms of Anionic Polymerization; Advances in Polymer Science; Springer-Verlag: West Berlin, 1983; Vol. 49. 8. The Poisson function, Nn/N = [ 1 /(n -1 )!]i/n' V l/, is a statistical model that applies to an ideal "living: polymerization in which there is no chain transfer or chain termination, and all active centers are growing at the same rate. For a discussion of the Poisson model for polymer size distribution, see: Flory, P. J. J. Am. Chem. Soc. 1940,62, 1561. 9. a) Watson, P. L ; HerskovKz, T. ACS Symp. Ser. 1983, No. 2 1 2 ,459. b) Watson, P. L J. Am. Chem. Soc. B 1982,104,337. 10. Jordan, R. F. J. Chem. Ed. 1988,65(4), 285. 11. This value for Ki corresponds to 30% phosphine dissociation (i.e, ca 0.02M free PMe3 ) in a 0.07M solution of the complex (Cp*SiNR)(PMea)ScR (the experimental concentration) in the absence of externally added PMea. 1 2. The numbers In parentheses beside the values for the parameters ki and Ki are the standard deviations determined in the non-linear, least-squares analysis. 13. This assumes an activation energy for olefin insertion of approximately 10 kcal mol*1 (Keii, T. In Transition Metals and Organometallics as Catalysts for Olefin Polymerization; Kaminsky, W.; Sinn, H., Eds.; Springer-Verlag: Berlin, 1988; p. 3. 14. Dividing the concentrations of added phosphine (L) in half in the non-linear, leastsquares analysis of the six data points increases the value for ki to 0.050(23)M*1sec*1 and decreases the value for Ki to 0.00011 (5) M*1. Halving the scandium concentration has a less significant effect, resulting in values for ki and Ki of 134 0.030(13)M'1sec'1 and 0.00075(33)M respectively. Likewise, reducing K2 by a factor of 2 has only a minor effect, resulting in a k^ of 0.034(14)M*1sec'1 and a K| of 0.00033(15)M. 15. Pino, P.; Mulhaupt, R. Angew.Chem. Int. E. Eng. 1 9 8 0 ,19,857 and references therein. 16. a) Corradini, P.; Busico, V.; Guerra, G. In Comprehensive Polymer Science', Allen, G., Bevington, J. C., Eds.; Pergamon Press: Oxford, 1989; Chapter 1 . b) Corradini, P.; Busico, V. Guerra, G. In Transition Metals and Organometallics as Catalysts for Olefin Polymerization; Kaminsky, W.; Sinn, H., Eds.; Springer-Verlag: Berlin, 1988; p. 337. 17. a)Bovey, F. A.; Jelinski, L W. Chain Structure and Conformation of Macromolecules; Academic Press: New York, 1982; Chapter 3. b) Bovey, F. A. In Comprehensive Polymer Science; Allen, G., Bevington, J. C., Eds.; Pergamon Press: Oxford, 1989; Chapter 17. c) Zambelli, A.; Dorman, D. E.; Brewster, A. I.; Bovey, F A Macromolecules 1973,6 (6), 925. d) Zambelli, A.; Locatelli, P.; Bajo, G.; Bovey, F. A. Macromolecules, 1975,8(5), 687. e) Ferro, D. R.; Zambelli, A.; Provasoli, P.; Locatelli, P.; Rigamonti, E. Macromolecules, 1 9 8 0 ,1 3 ,179. f) Toneili, A. E.; Schilling, F. C. Macromolecules 1 9 8 0 ,13 ,270. 18. Bunel, E. Ph.D. Thesis, California Institute of Technology, 1988. 19. a) Zambelli, A.; Gatti, G. Macromolecules, 1 9 7 8 ,11 (3), 485. b) Zambelli, A ; Grass), A.; Resconi, L; Albizzati, E.; Mazzocchi, R. Macromolecules, 1 9 8 8 ,2 1 ,617. 20. Kaminsky, W.; Kulper, K.; Brintzinger, H. H.; Wild, F. W. P. Angew. Chem. Int. Ed. Engl. 1 9 8 5 ,24, 507. 21. a) R0II, W.; Brintzinger, H. H.; Rieger, B.; Zolk, R. Angew. Chem., Int. Ed. Engl., 1990, 29(3), 279. b) Erker, G. Pure & Appl. Chem., 1989,67(10), 1715. 22. Ewen, J. A.; Jones, R. L ; Razavi, A J. Am. Chem. Soc., 1 9 8 8 ,110,6255. 23. a) Ref. 18 b) Ref. 5a c) An exception to this is the low-temperature polymerization of propene by Cp2 TI(Ph)2 , which yields isotactic polymer as a consequence of chain-end control. Ewen, J. A. J. Am. Chem. Soc. 1 9 8 4 ,106,6355. 24. Burger, B. J.; Bercaw, J. E. In New Developments in the Synthesis, Manipulation and Characterization of Organometallic Compounds; Wayda, A L , Darensbourg, M. Y. Eds.; ACS Symposium Series 357; American Chemical Society: Washinton, D. C., 1987; pp. 17-98. 25. Marvich, R. H.; Brintzinger, H. H. J. Am. Chem. Soc. 1 9 7 1 ,9 3 ,2046. 135 Part II Hydrazido(l-) and 2,2-Dimethylhydrazido(1-) Derivatives of Permethylscandocene. Preparation and Structural Characterization of Their Products from Reactions with Acetonitrile 136 INRODUCTION In the first part of this thesis we discussed the olefin-insertion activity of monocyclopentadienyl scandium amido complexes. This second part describes some work that was actually initiated prior to the olefin polymerization project and involves insertion chemistry with the permethylscandocene system. Besides olefins, there are a number of unsaturated substrates such as CO , 1 CO 2 , 2 nitriles, 3 alkynes4 and organic carbonyls,2 which undergo insertion chemistry with scandium hydride and alkyl derivatives. Because of the heteroatom affinity of the Lewis acidic, electronically unsaturated scandium center, oxygen and nitrogen containing substrates are especially reactive. The reaction chemistry between these small molecules and scandium has been explored primarily with the permethylscandocene system. A particularly interesting reaction of permethylscandocene hydride is its reduction of nitriles in the presence of H2 (Figure 1 ) . 3 N«OCM e, ,H CpJScH — ^ C p g S c ^ N -C ^ »C p N CMe3 HgN C HjC M ej HZ CpJScNHCH^CMej CpJSc NHjCHgCMej CHCMe / NHCH2CMe3 Figure 1 C p J S c *N « C CM«S Hg 137 The facility with which the scandium amide bonds are hydrogenated (leading to moderate catalysis for the hydrogenation of t-butylnitrile to neopentylamine) was surprising, since it had been assumed that early transition metal-nitrogen bonds were too inert to be involved in such a catalytic cycle. The recent discovery by Gagne and Marks® of a catalytic cycle involving the hydroamination/cyclization of amino olefins with organoianthides has further served to dispel this misconception. We were interested in exploring the possibility of extending the chemistry between permethylscandocene hydride and nitriles to the reduction of N2 (Equation 1). Cp*2ScH + N =N Cp#2S c = N = N / Hi ( 1) H, Cp*2S c = N H N H 2 The homogeneous, transition-metal promoted reduction of N2 with H2 has been a long­ standing goal of inorganic chemists that would have important commercial applications.6 The unfavorable thermodynamics for the partial reduction of N2 to either N2 H2 or N2 H4 7 posed a potential difficulty, since the hypothetical reactions shown in Equation 1 are also likely to be endothermic. 1-1 N2(g) + H2(g, N2H2(g) AH0 = + 43.5 kcal mol N2(g) + 2 H2(g) N 2H4(g) AH° - + 20.3 kcal mol*1 N2(g) + 3 H2(g) 2 NH3(g) 1-1 AH0 » -2 4 .9 kcal mol In order for the reaction to be productive, further reduction or disproportion of the Cp*2 Sc=NHNH2 species would be required (Equations 2 and 3). 138 2 Cp*2ScNH2 Cp*2ScNHNH2 + Cp*2ScH (2) ?H 2 2 Cp*2ScH + 2 NH3 ? 2 Cp*2ScNHNH2 Cp*2ScNH2 + Cp*2ScH + NH3 (3) N2 Therefore, we directed our efforts toward the synthesis of the permethylscandocene hydrazido(l-) complex to determine the possibility of promoting i-fe reduction of this species to ammonia or disproportion to N2 and NH3 . While the parent hydrazido(l-) derivative, Cp*2 ScNHNH2 , was straightforwardly prepared, the acidic nature of the N-H moieties makes this complex an unlikely intermediates in a catalytic process for N2 reduction. In addition, no productive hydrogenation of this species was observed. At the time of this work, unsubstituted hydrazido(l-) metal complexes were quite rare, presumably because of the instability of the highly reducing N2 H3 - ion.8 There are still relatively few reported examples of complexes having the unsubstituted hydrazido(l-) ligand, 8 two of which have been recently confirmed by x-ray structures.8 6 By contrast, there are several examples of fully characterized, alkyl and aryl substituted hydrazido(l-) species. 1 8 , 1 1 , 1 2 During our efforts (albeit unsuccessful) to obtain crystals of Cp*2 ScNHNH2 , we discovered an interesting reaction between this species and acetonitrile to form the fivemembered heterometallacycle, Cp*ScNHC(CH3)NNH2 . An X-ray structure determination confirmed the identitiy of this species. Our investigation of the mechanism by which this 139 heterometallacycle is formed is the focus of this chapter. As will be discussed, a 1,3-dipolar insertion of the acetonitrile into the Sc-N bond of the hydrazide derivative is the most likely mechanism for this reaction. The closely related compounds Cp^ScNHNMea and Cp*2 ScN(H)C(CH3 )NNMe2 also have been prepared and the crystal structure determination of the latter complex is reported as well. Results and Discussion Preparation Cp*2 ScNHNH2 . The synthesis of permethylscandocene amide complexes according to Eq. 4 has been previously reported. 3 Cp*2 ScCH3 + HNRR* —---------- 1> Cp*2 ScNRR‘ + CH4 (4) We find this method useful for the synthesis of scandium hydrazido(1*) derivatives as well (Eq. 5). Cp*2 ScCH3 ♦ H2 NNRR’ -------------> Cp*2 ScNHNRR' + CH4 1: R = R' = H 2: R = R’ = CH3 (5) Although the reaction between Cp*2 ScCH3 and one equivalent of NH2 NH2 proceeds cleanly (1H NMR), excess hydrazine protiolytically cleaves the [Cp*-Sc] bonds, resulting in decomposition accompanied by loss of Cp*H (1H NMR). Cp*2 ScNHNH2 (1) appears to be indefinitely stable in the solid state at 25* C, but it decomposes in benzene solution over several days, again with the appearance of free Cp*H (1H NMR), likely the result of (sterically hindered, thus slower) intermolecular protiolytic cleavage of [Sc-Cp*] bonds by [SC-NHNH2 ]. The room temperature, 400 MHz 1H NMR spectrum for an unpurified sample of 1 displays an average signal for all three [NHNH2 ] protons, which separates into two (still rather broad) signals of 2:1 relative intensity upon cooling to -96 *C. A recrystallized sample, however, displays the two separate hydrazide proton resonances at room temperature. Presumably, trace amounts of residual hydrazine catalyze proton exchange between the two types of 140 hydrazide sites, analogous to the base-catalyzed hydrazide(l-) proton exchange noted earlier for Cp*W(CH3 )3 (»j2-NHNH2 ).9c Infrared, 1H NMR, mass spectral and analytical data for 1 are given in the Experimental Section. The spectral data do not, however, distinguish between Cp*2 Sc(rj1-NHNH2 ) and Cp*2 Sc(»j2-NHNH2 ) structures (or some other fluxional alternative with an N-H agostic interaction19) for 1. Because of the novelty of its formulation (particularly so at the time of its first isolation), many attempts were made to obtain crystals of Cp*2 ScNHNH2 suitable for an x-ray structural determination; unfortunately all such attempts have been unsuccessful. In view of the dihapto coordination found by Schrock and coworkers for [Cp*W(CH3)3(»j2-NHNH2 )] ♦ and Cp*W(CH3)4(»f2-NHNH2 )9c and the 0 C-H agostic structure we have observed for Cp*2 ScCH2 CH3 , 1 4 we favor the dihapto structure for 1. Reaction of Cp*2 ScNHNH2 with Acetonitrlle. Upon addition of excess acetonitrile to 1 in benzene at 2S*C, clear crystals are deposited from solution within minutes. Incorporation of one molecule of acetonitrile in this new compound (Eq. 6) is indicated by 1H and 13C NMR data; however, an IR stretch in the region 2100-2300 cm'1 diagnostic of a O N group (attributable, for example, to a simple acetonitrile adduct, Cp*2 Sc(N-CCH3)(*r1-NHNH2 )) is absent Cp*2 ScNHNH2 + CH3C-N 1 ---------- 1> Cp*2 ScN(H)C(CH3)Nrta2 3 (6) Lower frequency vibrations at 1591 cm'1 and 1607 cm'1, not present for 1, are observed, however. A single-crystal x-ray structure analysis was then undertaken, and the ORTEP drawing of the molecular structure is shown in Figure 1. As is apparent the acetonitrile and hydrazido(l-) moieties have combined, forming a five-membered heterometallacycle, Cp*2 Sc-NH-C(CH3 )-N-f$H2 (3). The planarity of the ring (to within ±0.12A) and nearly equivalent C1-N2 and C1-N3 bond lengths suggest resonance stabilization in the ring, with the two canonical forms shown in Scheme 1 (vido infra) being the principal contributors. The Me4 141 3 Figure 2. ORTEP drawing of hydrogen bonded Cp*2ScN(H)C(CH3)NNH2 pair. 142 143 N1 -N 2 bond length of 1.483(5)A is rather long in comparison to that found for other hydrazido(l-) complexes1 0 , 1 1 , 1 2 but comparable to that found for N2 H4 . 1 5 An interesting feature of the crystal structure is the N1-N1 H b -N 2 hydrogen bonding between two centrosymmetrically related molecules, as evidenced by the short /nfermolecular N1-N2 distance of 3.08(5)A. An ORTEP drawing showing the hydrogenbonded pair of molecules is given in Figure 2. Involvement of hydrazide protons in hydrogen bonding has been observed in a variety of metal hydrazide complexes9 ® ’ 1 ° ® * 1 0 c ’ 1 1 The mechanism for the formation of 3 from 1 and CH 3 CN is of some interest. Since the acetonitrile unit could be oriented in either the C2C1-N2 or C2C1-N3 position, a number of possible mechanisms for the formation of 3 present themselves (Schemes 1 and 2). The most direct pathway is mechanism A (Scheme 1), involving cycloaddition of acetonitrile with NH-NH 2 bond cleavage. The alternative direction of cycloaddition (mechanism B) would initially yield a tautomer of 3, which would need to undergo a subsequent hydrogen shift from one a N atom to the other to generate 3. While the two cycloaddition mechanisms shown in Scheme 1 place the acetonitrile in positions C2C1-N2,1® two (more plausible) mechanisms that place the acetonitrile in the C2C1-N3 position must also be considered (Scheme 2 ). Mechanism C involves 1,3 migration of the hydrazide (1-) group to the carbon atom of coordinated acetonitrile, analogous to the reactivity of permethylscandocene alkyl, aryl and hydride derivatives with nitrites (Eq. 7)3 , followed by tautomerization to 3. Cp*2 So-R + R 'C -N e> C p*2 Sc=N=CRR’ (7) (R = H, CH3, CeHs; R' = CeH5, CMe3) Finally, the distinguishing feature of mechanism 0 is intramolecular nucleophilic attack by the terminal nitrogen of the hydrazide group at the carbon atom of the coordinated nitrite, a proposal of some merit in light of the generally observed activation of unsaturated ligands 144 Scheme 1 mechanism A mechanism B H CHj 6 / I Cp'jSc' || nh '" a \ J ftV : :N ^ ^N H a C p -jS c ^ / H H ShHt H :N . C p'aSc/ \ H' k \ N; .x C H * C p'aS d^ I N" h '" ,CHa s? c' k H jH i 145 Scheme 2 mechanism C CH, 1,3 NHNHj migration -------------- J *N * ^Ha CH3C15W Cp'aSc Cp*2Sc CH3 _ / C p*jSc= ,SN — C NH— NHa NH NH— NH2 coordinate i -NHj H H \ \ .C H j : ,S/V- / / ^ C c x CH3 - ,5^ c p -V Cp-jSc, Cp'jSc, \ N" n n: V ..N h" i H H'" I H H'" i H I \ \ N ^— •• : n: ^ ch3 H ,C H , nudeophilo attack by-NK] o p .^ /r « \ A.. ^ 2 . W a< ! Cp'aSc; \ MM M l NH— NH* _ ^ N —.||J I \ H H H shift H H x ch3 = ; ^ C p -V || — — \ . c- ch. Ja C p 'aS o \^ - N A •N H , O p-aS./ | H' k H : " \ .N H" ^ H | N: v H 146 toward nucleophilic attack by coordination to an electron-deficient metal. 1 7 The requirement to move both hydrogens from N2, one to N1 and one to N3, to generate 3 is an unattractive feature of this alternative, however. A labeling experiment carried out with 1 5 N«CCH 3 revealed that the acetonitrile nitrogen finally resides at N3, apparent from the sharper proton resonance at 6 3.05, with 1J i5N-H = 70 Hz attributable to Cp*2 Sc1 5 N(H)C(CH 3 )NNH 2 . Hence, mechanisms A and B, and for that matter any mechanism that positions the acetonitrile nitrogen other than at N3, may be discarded. Reaction of C p*2 ScNHNM » 2 w ith A cetonitrile. To decide between the remaining two mechanisms, we examined the reaction of acetonitrile with Cp*2 ScNHNMe2 (2). We reasoned that by replacing the two hydrogens on the terminal nitrogen of the hydrazide with two methyl groups, formation of a complex analogous to 3 could not occur by mechanism D, since the necessary hydrogen shift would no longer be possible. Reaction of 2 with 15N labeled acetonitrile resulted in a new species that displayed 1 5 N-1H coupling in the 1H NMR (6 3.9 w ith 1 Jisn-h = 73 Hz), indicating that a product analogous to 3 had formed. Moreover, infrared data also indicated that the acetonitrile once again had reacted beyond simple adduct formation (Eq. 8 ). Cp*2 ScNHNMe2 + CH 3 C -N 2 ----------- > Cp*2 ScN(H)C(CH 3 )NNMe2 4 (8 ) Ultimately, positive structural verification was provided by a single-crystal x-ray analysis for 4 (Figure 3). A four-membered metallacycle is formed in this case. The. smaller ring size adopted by 4 relative to 3 is undoubtedly due to the steric bulk of the dimethylamino group, which prevents coordination to the [Cp*2 Sc] moiety. Resonance stabilization attributable to the two canonical forms shown for 4 in Scheme 3 is apparent from the similarity of the two C-N bond lengths (C1-N31.306(7)A; C 1-N 21.316(7)A). Hydrogen bonding between monomers, analogous to that observed in the solid state for 3, is not expected and is not 147 Z | ri O £> 21 CM a$ ■o rt 0 . (A LU 2 cn s o * Me 1 0 U. h* ro u , 148 Scheme 3 Cp#2S c -N H N (C H 3)a CH3C ,5/V CHj 1,3 NHN(CHj), J5N 4 Cp’ jSc migration -C NH— N(CHj)a \ pH j coordinate C p *a S c = ,9 W * q — / Cp*aSc \ i H —N(CHj)j \ C — CH3 V / \ n: h CHj *t:H j - 0 nudeopNHc attack by -N(CHj)i H Shift H H ^ CHa / Cp'aSe, ^ C p 'jS c. C— C H jn: A cp ,jSc\ / , _ C H )I N \ n: n: CyH , \ * . CH^ 'c h i J 149 observed for 4. The implications from the reaction of 2 with acetonitrile (Scheme 3) concerning the mechanism for the formation of 3 (mechanism C vs. D, Scheme 2) are not, however, definitive. Although one may safely infer that mechanism C is plausible, mechanism 0 is not ruled out conclusively. When the 0 nitrogen of the hydrazido(l-) ligand does bear two hydrogens, the reaction with acetonitrile may choose to follow pathway D rather than C. The balance of evidence does argue for mechanism C, nonetheless. Insertion of a C«N bond into a scandium-nitrogen bond has been observed before as a competing reaction in the catalytic hydrogenation of fert-butylnitrile to fert-butylamine by permethylscandocene hydride. 8 In some related chemistry of amide derivatives of aluminum, nitrile insertion has been observed to occur at the Al-N bond in preference to the Al-C bond. 1 8 Moreover, the insertion of a nonpolar unsaturated group into a transition metal-nitrogen bond is a key step in the catalytic hydroamination/cyclization of a.wamino olefins,8 as mentioned earlier, as well as in the recently reported formation of 2,3diazametallacyclopentenes from the reaction of Cp2 Zr(N 2 Ph2 ) with internal alkynes. 1 8 The tautomerization of the initial product of the nitrile insertion, via a formal 1,3 hydrogen shift (mechanism 0 , Scheme 2 ), bears some resemblance to the analogous process observed for the insertion product of Cp*2Ti(»?2-C2H4) acetonitrile, which yields Cp*2^CH2CH=C(CH3)dH.20 Summary and Concluaiona The modest stability of these hydrazido(l-) derivatives of permethylscandocene, C p*2 ScNHNH 2 and C p* 2 ScNHN(CH 3 )2 , may be traced, in part, to the very high reduction potential of these formally Sc(lll) complexes, although in view of the moderate stability of the formally W(VI) complexes [Cp*W (CH3)3(NHNH2) r and Cp*W (CH3)4(NHNH 2 ),9b*c one may question the premise that the high reduction potential for [N 2 H3]' is a principal reason for the 150 paucity of examples of stable transition metal derivatives with the hydrazido(l-) ligand. The susceptibility of the [Sc-Cp*] bonds to protiolytic cleavage limits the stability, especially for Cp*2 ScNHNH 2 , and restricts the types of chemistry that may be explored. For example, a clean reaction with acetone to yield [Cp*2 ScNHN=C(CH 3 )2 ] is unlikely, since water (the co­ product) reacts rapidly with all permethylscandocene compounds we have prepared to date. On the other hand, the insertion chemistry observed with acetonitrile is particularly clean, and may provide an efficient means to trap less stable adducts in other systems. Table I. 1H and 13C data for Cp*2 Sc Complexes. 151 Table I. 1H and 13C data for Cp*2 Sc Complexes (continued). 152 153 T a b le I I . C r y s ta l a n d In te n s ity C o lle c tio n D a ta fo r C p 5 S c N (H )C (C H « ) N N H a Form ula: ScN j C jj H ss Form ula weight: 387.53 C rystal size: (estim ated.) 0.7 x 0.3x < 0.5 m m C rystal color: Colorless Space group: P2i/n #14 Absences: OfcO, Jfe odd; hOl, h +■ / odd a = 9.50fl(3)A 0 ■ 101.30(1)* b = 14.742(2)A c = 15.608(2)A V * 2177.0(0)A* Z = 4 H = 3.09 c m '1 PttU —1.139 gem -1 C A D -4 diffractom eter tti scan A = 0.71073 A G raphite m onochrom ator 29 range: 2 * -2 0 * O ctants collected: —ht ± k,±l 29 range: 2 0 *-2 5 * Octants collected: —h, - k , ±1 T = 294*K Num ber o f reflections measured: 0734 Num ber o f independent reflect ions: 3855 Num ber w ith FJ > 0: 3278 Num ber w ith F* > 3#(F*): 2004 Goodness o f flt fo r m erging d ata: 0.945 R fo r m erging duplicate reflections: 0.040 N um ber o f reflections used in refinem ent: 2207 F in a l R -index: 0.117 (0.051 fo r F * > 3 * (F » )) F in a l goodness o f fit: 1.34 for 380 param eters 154 T a b le H I. N o n -H y d ro g e n A to m C o o rd in a te s a n d D is p la c e m e n t P a ra m e te rs fo r C p jS c N (H )C (C H * ) N N H a *»y, x and U,n* x 1 0 « Atom X y z Sc 3113(.8) 2326(.4) 359(.5) 337(2) Nl 4588(4) 1152(3) 65(2) 421(11) N2 4191(4) 662(2) -739(2) 471(10) N3 2168(4) 1543(3) -6 6 0 (2 ) 462(11) Cl 2925(5) 914(3) -1 0 1 9 (3 ) 427(12) C2 2421(10) 438(6) -1 8 3 0 (5 ) 692(20) Cpl 2796(4) 2455(3) 1952(2) 402(11) Cp2 1472(4) 2677(3) 1552(2) 425(11) Cp3 972(4) 1890(3) 1136(2) 434(12) C p4 1935(4) 1180(3) 1284(2) 407(11) Cp5 3068(4) 1543(3) 1785(2) 390(11) Cp6 3731(5) 4001(2) 383(3) 438(12) C p7 5007(4) 3517(2) 2 93 (3 ) 395(11) Cp8 4947(4) 3109(2) -5 1 9 (3 ) 394(11) C p9 3641(8) 3308(3) -9 2 6 (3 ) 430(12) C plO 2886(4) 3845(2) -3 6 5 (3 ) 421(11) M el 3658(8) 3006(5) 2591(4) 673(19) M e2 844(7) 3527(4) 1666(5) 691(18) M e3 -4 * 4 ( 6 ) 1798(5) 6 51 (5 ) 710(20) M et 1746(6) 227(4) 989(5) 672(19) 155 T t b l e m . (C o n t.) Atom z y Me5 4299(7) 1002(5) 2162(4) 647(17) Me 6 3497(10) 4716(4) 1045(5) 726(20) Me7 6279(7) 3554(6) 904(5) 746(19) Me8 6150(6) 2660(4) -951(4) 661(16) Me9 3249(10) 3083(5) -1833(4) 743(21) MelO 1431(7) 4241(5) -583(5) * 761(20) T a b le ltv. H y d ro g e n A to m C o o rd in a te * e n d D is p la c e m e n t P a ra m e te rs fo r C P ; S c N ( H ) C ( C H ,) N N H a x,y and x x 104 A tom X y X N lH B N lH A N 3H C 2H A C2HB C 2H C H 1A H lB H 1C H 2A H2B H 2C H 3A H3B H 3C H 4A H 4B H 4C HSA B 5B H 5C H 6A H 6B H 6C H 7A H 7B H 7C HSA H 8B H 8C HSA H 9B H 9C H 10A H10B H lO C 4719(39) 3343(36) 1501(32) 2678(66) 1779(50) 2144(47) 3332(49) 3748(36) 4281(56) 885(50) 573(53) -1 5 6 (4 7 ) -4 9 3 (5 6 ) -1 1 8 2 (5 2 ) -6 5 0 (4 8 ) 2483(50) 1374(45) 1540(50) 4980(47) 4178(42) 4452(50) 4047(45) 2742(70) 3639(58) 6138(56) 6380(56) 6995(59) 6760(42) 5864(46) 6485(55) 3281(54) 2356(57) 3434(46) 1512(43) 749(64) 1308(49) 662(23) 1390(22) 1608(23) -9 9 (3 9 ) 199(37) 910(29) 3532(31) 2714(22) 3305(38) 3826(29) 3911(31) 3395(30) 1238(33) 1749(31) 2312(26) -6 9 (3 2 ) -6 6 (2 9 ) 119(30) 1369(29) 853(26) 376(30) 5190(30) 5031(47) 4500(31) 3992(34) 3021(34) 3758(37) 2440(26) 2062(28) 3049(33) 2474(32) 3074(39) 3464(27) 4653(26) 3736(38) 4643(27) 384(22) -3 1 (2 2 ) -8 8 2 (2 0 ) -1 9 3 9 (3 4 ) -1 7 1 1 (3 4 ) -2 2 2 4 (2 8 ) 2746(28) 3130(23) 2388(33) 2169(29) 1150(32) 1816(30) 283(32) 1014(29) 328(25) 961(29) 1391(25) 394(27) 2095(28) 2744(25) 1890(27) 1033(26) 920(41) 1602(29) 1337(33) 1183(33) 658(34) -5 3 0 (2 4 ) -1 2 7 6 (2 6 ) -1 3 4 8 (3 2 ) -1 9 5 8 (2 9 ) -1 9 8 5 (3 7 ) -2 1 9 6 (2 7 ) -1 0 3 8 (2 4 ) -6 8 5 (3 5 ) -1 5 4 (2 7 ) B 4 .5 (1 2 )* 2.6(11)* 1.5 (1 0 )* 11.2(23)* 6 .3 (2 2 )* 7 .7 (1 6 )* 6 .4 (1 6 )* 4 .4 (1 0 )* 9 .4 (2 1 )* 7 .7 (1 7 )* 9 .8 (1 8 )* 7 .0 (1 7 )* 1 1.4 (2 1)* 8 .9 (1 7 )* 7 .0 (1 5 )* 6 .8 (1 9 )* 5 .8 (1 5 )* 8 .4 (1 7 )* 6 .2 (1 6 )* 6 .2 (1 3 )* 9 .0 (1 7 )* 6 .2 (1 4 )* 15.2 (3 3)* 8 .7 (1 8 )* 1 0.5 (2 1)* 1 0.7 (2 2)* 10.8 (2 3)* 5 .4 (1 3 )* 7 .9 (1 5 )* 1 0.5 (2 0)* 8 .5 (1 8 )* 1 0 .5 (2 5 )* 5 .6 (1 4 )* 5 .9 (1 3 )* 1 3 .1 (2 5 )* 6 .7 (1 7 )* 157 Table V . Aniaotropic Therm al Dlsplacpment Parameters for C p 5 S c N (H )C (C H ,)N N H a u„ A tom Sc Cpl Cp7 Cp8 N2 N3 Cp2 N1 M e9 Cp3 Cp9 M e8 Cp4 Cp6 M e6 Cp5 M e2 M e7 M elO C2 M el C plO Cl M e4 M e5 M e3 298 (4) 413 26) 351 27 367 28 1 551 26 395 28 421 28 416 28 1026 63 305 27 4951 30 635 39 432 28 557 31 10481 56 4091 28 1 622 43 1 622 '44' 597 43 787 54 8501 48 3691 27 1 559( 33 7954 51 7221 45 347(34] 328(4) 460 [31 357 25 354 24 451 22 514) 25 356 24 358 25 6831 47 512 28 4184 27 6401 37 3244 26 3134 25 372 34 4041 26 6354 40 9254 52 7104 46 677 48 767 50 341 25 3434 26 481 38 741 47 8664 47 13 385(4) 340(23) 481(28) 469(28) 407(22) 465(26) 509(26) 486(26) 508(38) 490(27) 377(27) 7 3 1 (4 l) 467(27) 459(27) 785(47) 361(25) 843j 50 681 '44' 977 54 592 42 409< 35 5531 29 3751 27 7431 51 4701 34 914(47' Uitj valoea h a re b e ta m ultiplied b j 104 T h e form o f th e displacem ent factor is: exp - 2 * * ( l r n h V + C Tn^k** + £T,af*c-' + 2Uu h ka T + 2Cr,j W a 'e * + 2 !fa A < » V ") U 3i -1 (4 ) -3 0 (2 1 ) 13(21) 2 5(21 ) -1 2 0 (1 8 ) -6 6 (1 9 ) -1 3 (2 3 ) 12(21) 1 66 (3 5 ) 4 4 (2 3 ) 5 4 (2 1 ) 6 5 (3 7 ) 1 6 (2 1 ) —2 9 (2 1 ) -1 3 7 (3 2 ) 6 4 ( 2 l) 4 6 (3 8 ) 3 0 (4 2 ) 226 4 4 ) -1 5 3 (3 6 ) -1 1 5 ( 3 3 ) 7 9 (2 3 ) -3 5 (2 l) 6 1 (3 5 ) 1 8 3 (3 3 ) 1 2 6 (4 3 ) T a b le V I. C o m p le te D is ta n c e s a n d A n g les fo r C p J S c N (H )C (C H * ) N N H , Distaace(A) Sc Sc Sc Sc -N 3 -N l -C e n l -C en 2 Cpl -C p 2 -C p S Cpl -M e l Cpl -C p S Cp7 Cp7 —CpS -M e 7 Cp7 Cp7 -C e a 2 CpS -C p S Cp8 -M e S -N 1 N2 N2 -C l -C l N3 N3 -N 3 H -C p S Cp2 Cp2 -M e S N1 -N 1 H B -N lH A N1 Me® -C p S Cp3 -C p S Cp3 -M e S CpS -C p lO Cp4 -C p S Cp4 -M e S CpS -M e S CpS -C p lO CpS -M e S M ^ O -C p lO C2 -C l N 1H B - N lH A 2.132(4) 2.287(4) 2.210 2.236 1.419(5) 1.395(5) 1.503(8) 1.403(5) 1.419(5) 1.508(8) 1.205 1.401(5) 1.505(7) 1.483(5) 1.316(5) 1.315(6) 0 .7 2 (3 ) 1.403(6) 1.494(8) 0 .8 8 (4 ) 0.8 2 (3 ) 1.488(9) 1.402(6) 1.513(8) 1.401(6) 1.410(8) 1.487(8) 1.409(8) 1.407(6) 1.515(7) 1.527(8) 1.507(9) 1.3 9 (8 ) Angle(°) C en l -Sc -C en2 Cp5 -C p l -C p 2 M el -C p l -C p 2 M el -C p l -C p 5 -C p 7 -C p 8 CpS M e7 -C p 7 -C p 8 M e7 -C p 7 -C p 6 CpS -C p S -C p 7 MeS -C p S -C p 7 MeS -C p S -C pS Cl -N 2 -N 1 -C l N 3H -N 3 CpS -C p 2 -C p l -C p 2 -C p l M e2 M e2 -C p 2 -C p 3 N 1H B -N 1 -N 2 -N 2 N lH A -N 1 -N 1 H B N lH A -N 1 -C pS -C p 2 Cp4 MeS -C p S -C p 2 MeS —CpS —CpS MeS -C p S —CpS C plO -C p S -C pS C plO -C p S -M eS CpS -C p S —CpS -C p S -C pS MeS MeS -C p S -C pS MeS -C p S -C p 7 C plO —CpS -C p 7 C plO -C p S -M eS CpS -C p S —C p l —CpS —C p l MeS —CpS —CpS MeS —C plO —CpS CpS M elO —C plO —CpS M e lO —C plO —CpS -N 2 N3 -C l C2 -N 2 -C l -N 3 C2 -C l 138.5 1 0 8 .0 (3 ) 1 2 7 .9 (4 ) 1 2 3 .0 (4 ) 1 0 7 .6 (3 ) 1 2 6 .1 (4 ) 125.7(4) 108.7(3) 126.2(4) 124.4(4) 109.9(3) 112.2(26) 106.8(3) 127.1(4) 125.4(4) 95.9(23) 104.4(24) 110.3(33) 109.6(3) 125.4(4) 124.9(4) 124.5(4) 107.8(3) 127.4(4) 106.7(3) 125.9(4) 127.4(4) 125.0(4) 107.2(3) 126.2(4) 109.0(3) 125.6(4) 125.2(4) 108.8(3) 124.1(4) 127.0(4) 124.5(4) lifc 3 (4 ) 121.8(5) 159 T a b le V I I . C r y s ta l a n d In te n s ity C o lle c tio n D a ta fo r C p J S c N (H )C (C H s )N N (C H s )a Formula: SCN3 C 2 4 H 4 0 Formula weight: 415.56 Crystal color: Colorless Habit: Prismatic Crystal size: 0.11 x 0.37 x 0.64 mm Space group: P 2 i/n #14 Absences: 0Jle0,fc odd; hQl,h -r I odd a = 10.953(2)A b = 19.733(1)A 0 3 101.30(1)° c = 11.441(1)A V = 2 4 2 4 .8 (5 )A * Z 34 fi 3 3.24 cm **1 (/iT n u =* 0 .1 2 ) P ttU * 1-138 g cm - * C A D -4 diflrsctom eter u> scan A * 0.71073 A G raphite monochromator 29 range: 2 ° -4 0 ° O ctants collected: ± h, k, ±1 T 3 2 94 *K Num ber o f reflections measured: 4873 N um ber o f independent reflections: 2267 N um ber w ith PJ > 0: 2089 N um ber w ith FJ > 3 w (F j): 1540 Goodness o f fit fo r m erging data: 1.08 R for m erging duplicate reflections: 0.0368 Num ber o f reflections used in refinem ent: 2267 F in a l R -index: 0.0640 (0.0416 fo r F * > 3 < r(F j)) F in a l goodness o f flt: 1.78 fo r 253 param eters 160 T a b le V m . N o n -H y d ro g e n A to m C o o rd in a te s a n d D is p la c e m e n t P a ra m e te rs fo r C p jS c N (H )C (C H s )N N (C H « )3 x ,y , z and Ut%* x 10* A tom X y X Sc 8390(.7) 1456(.4) 7213(.7) 390(2) Cpl 10255(4) 2235(2) 7708(4) 432(13 Cp2 9191(4) 2629(2) 7718(4) 412(13 Cp3 8672(4) 2427(2) 8689(4) 427(13 Cp3 9391(5) 1898(2) 9270(4) 463(14 CpS 10372(4) 1778(2) 8665(4) 486(15 M el 11218(4) 2396(3) 6972(4) 716(16 M e2 8757(5) 3208(3) 6898(4) 703(16 MeS 7556(5) 2759(2) 9045(4) 712(16 M e4 9265(5) 1579(3) 10434(4) 830(19 MeS 11444(5) 1304(3) 9107(4) 841(18 CpS 8881(4) 1176(2) 5204(4) 438(13 Cp7 9713(4) 788(2) 6021(4) 465(14 CpS 9013(4) 318(2) 6543(4) 446(14 CpS 7750(4) 394(2) 5994(4) 448(14 C plO 7675(4) 925(2) 5171(4) 438(14 MeS 9189(4) 1714(3) 4357(4) 707(16 M e7 11110(4) 744(3) 6111(4) 727(17 MeS 9544(5) -2 4 3 (3 ) 7387(5) 748(17 MeS 6687(5) -7 2 (3 ) 6089(4) 673(16 N1 6166(4) 702(2) 8743(4) 775(15 N2 6758(4) 1164(2) 8055(4) 583(13 N3 6592(4) 1925(2) 6619(4) 632(14 161 Table v m . Cl 6017(4) 1516(3) 7230(6) 664(20) C2 4611(5) 1451(3) 6998(6) 1175(27) C3 7028(6) 162(3) 9191(5) 890(20) C4 5807(6) 1050(3) 9755(6) 1227(23) au* i - \ E i • *>) I 162 Table IX . Assigned Hydrogen A tom Coordinates and Displacement Parameters for C p jS c N (H )C (C H i)N N (C H *)a x ,y and z x 104 Atom X y z B H la H lb H lc H 2a H2b H2c H 3a H3b H3c H 4a H 4b H4c H 5a H5b H5c H fla H 6b Hflc H 7a H7b H7c H 8a H8b H8c H fla H flb Hflc H lO a H lO b H lO c HN3 H C 2a H C 2b H C 2c H G 3a H C 3b HCSc H C 4a H C 4b HC4c 11681 10803 11740 9374 7996 8635 7316 6906 7786 9374 8466 9895 12094 11157 11699 10067 8851 8830 11345 11299 11501 9572 9024 10358 6300 6108 7001 6538 5816 6533 6313 4294 4299 4402 7258 6619 7731 5397 5249 6526 2774 2490 2010 3548 3369 3042 2488 2776 3197 1105 1682 1763 1407 852 1368 1739 2128 1577 1071 300 828 -6 4 4 -3 0 2 -1 1 3 -2 1 2 , 172 -4 4 9 1599 993 895 2295 1563 1761 1002 -5 9 -1 4 4 356 732 1406 1222 7320 6188 6993 7020 7062 6097 9653 8367 9335 10369 10578 11042 8700 8934 9935 4456 4546 3568 5592 5877 6910 6937 7949 7773 5313 6453 6565 4181 4600 3561 6101 6187 7507 7162 8534 9629 9694 10164 9439 10244 7.2 7.2 7.2 7.1 7.1 7.1 6.8 6.8 6.8 8.0 8.0 8.0 7.9 7.9 7.9 6.7 6.7 6.7 6.8 6.8 6.8 7.3 7.3 7.3 6.6 6.6 6.6 7.1 7.1 7.1 5.9 10.4 10.4 10.4 8.2 8.2 8.2 10.5 10.5 10.5 ‘ Isotropic displacem ent param eter, B 163 Table X. A n is o tro p ic T h e r m a l D is p la c e m e n t P a ra m e te rs x 104 fo r C p ; S c N (H )C (C H « ) N N (C H .)a A to m U\ i CTjj U:33 Sc 339(5) 410(32 471(32 441(31 590(35 351(31 571(34 824(41 752(38 1156(47 785(40 474(33 355(32 447(33 469(35 395(34 765(39 466(36 813(41 673(37 726(34 495(29 352(5) 444(34 328(31 368(32 406(33) 45l(34 814(42 535(36 606(39 754(44 848(47 458(33 486(35 342(31 337(33 453(35 768(42 864(43 543(38 565(38 597(35 482(28 477(30 587(42 1169(55 722(47, 1086(54) 468(6) 455(34 423(33 487(32 362(32 577(36 806(40 738(40 852(42 543(36 733(39 384(31 544(34 505(33 528(33 43l(32 609(36 852(41 828(41 Cpl Cp2 Cp3 Cp3 Cp5 M el Me2 Me3 Me4 MeS Cpfl Cp7 Cp8 Cp9 CplO Me6 Me7 MeS Me® Nl N2 N3 Cl C2 C3 C4 534(29 356(35 397(35 1074(52 1372(59 7 7 0 (4 0 ' 1134(42 822(35 846(35 1025(49 1907(64' 980(49' 1552(66 Vl 3 13 49(4) 120(26 52(26 13l(27 17(27 -105(27 243(30 126(31 336(32 79(32 -234(32 90(26 62(27 -17(27 76(27 -3 (2 5 187(30 131 30 1 0 <33 l l l i 30 5041 32 2491 26 38 25 74 34 1 0 O 38 4 6 » 41 1092 54 Ui J -26(5) -109(27) 6 25) —78(26) 23(26) —85(29) -278(32) 59(31) -225(32) 60(32) -205(33) -43(26) -163(28) -32(26) -78(27) -49(26) -61(31) -336(34) -66(31) -93(29) -148(31) -67(25) 12(25) -259(39) -3 0 8(5 l) -53(38) -267(50) T h e fo rm of th e displacement factor is: exp - 2 * * ( U u / k V ' + V + !7u J V a + 2Ul2hka-b* + 2 £ 7 „M a 'c * + 2Unhib*c*) 164 T a b le X I . C o m p le te D is ta n c e s a n d A n g le s fo r C p 5S c N ( H ) C ( C H 8 ) N N ( C H , ) , Oistance(A) Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc Sc -N2 -N3 -C pA -C pB -C p l -Cp2 -Cp3 -Cp4 -Cp5 -C p 6 -C p 7 -C p 8 -C p 9 -C p lO N l -N 2 N1 -C 3 N1 -C 4 N2 - C l N3 - C l C l -C 2 C p l -C p 2 C p l -C p 5 C p l -M e l C p2 -C p 3 C p2 -M e 2 C p3 -C p 4 C p3 -M eS C p4 -C p 5 C p4 -M e 4 C p5 -M e S C pfl -C p 7 CpS -C p lO Cpd -M eS C p7 -C p S C p 7 -M c 7 CpS -C p S CpS -M e S CpS -C p lO CpS -M eS C p lO -M e lO 2 .2 6 4 2.161I 2 .2 3 3 2 .2 3 1 2.531 2.502i 2.533i 2.5471 2.5421 2.52 2 .5 4 2.5121 2.53 2.5421 1.43 1.451 1.46 1.318 1.3061 1.5161 1.4031 1.4051 1.507 1.401 1.4SSI 1.383 1.511i 1.407 1 .5 0 « 1.5101 1.3SSI 1.4051 1.518 1.408 1.518 1.411 1.508 1 .4 0 8 1.5041 1.508 Angle(s) CpA -Sc CpA -Sc CpA -Sc CpB -Sc CpB -Sc N3 -Sc Sc -N3 Sc -N 2 C3 - N l C4 - N l -CpB -N2 -N3 -N2 -N3 -N2 -C l -C l -N 2 -N 2 C4 - N l —C3 Cl -N 2 - N l N3 - C l -N 2 C2 - C l -N 2 C2 - C l -N 3 CpS -C p l -C p S M e l —C p l —CpS M e l —C p l —Cp5 Cp3 -C p S -C p l MeS -C p 2 -C p l MeS -C p S -C pS C p4 -C p 3 -C p S MeS —CpS —CpS MeS -C p S -C p 4 CpS —C p 4 —CpS MeS —CpS —CpS MeS —CpS —CpS CpS —CpS —C p l MeS -C p S -C p l MeS -C p S -C p S C p lO -C p S -C p 7 MeS -C p S -C p 7 MeS -C p S -C p lO CpS -C p 7 -C p S M e7 -C p 7 -C p S M e7 -C p 7 -C p S CpO -C p S -C p 7 MeS -C p S -C p 7 M e * -cps -cpo C p lO -C p S -C p S MeO -C pO -C p S MeO -C pO -C p lO CpO -C p lO -C p S M e lO -C p lO -C p 6 M elO -C p lO -C p O 137.8 108.5 107.4 110.6 105.6 56.8 9 4.2 8 8 .3 109.21 110.8 108.8 116.5 114.6 122.8 122.6 107.3 124.2 127.2 108.4 125.4 125.9 108.2 124.5< 127.3< 107.8 126.3 125.1 105.3 127.2 124.3 108.1 127. 123. 107. 126. 124.3 108.3 125.3 125.7 107. 127. I2 4 .li 108. 125.3 125. 165 EXPERIMENTAL SECTION General Considerations. All manipulations wars performed using glovebox and high-vacuum techniques as described elsewhere*2 1 Argon, hydrogen and nitrogen were purified by passage over MnO on vermiculite and activated 4A molecular sieves. Solvents were dried and purified by prolonged reflux over a suitable drying agent, followed by distillation under an atmosphere of dinitrogen. Ether solvents were stored over sodium benzophenone ketyl; hydrocarbon solvents were stored overtitanocene; 2 2 and methylene chloride and acetonitrile were stored over CaH 2 . Anhydrous hydrazine (Aldrich) was transferred to a Kontes microflex vial with rotary-type valve and used without prior purification. 1,1-Dimethylhydrazine (Aldrich) was stored over 4A molecular sieves in a flask under vacuum and used as received. NMR spectra were recorded on Varian EM-390 ("lH, 90MHz), JEOL FX90Q (1 H, 89.56 MHz; 1 3 C, 22.50MHz) and JEOL GX400Q (1 H, 399.78 MHz) instruments. Infrared spectra were measured as Nujol mulls on KBr plates and recorded on a Beckman IR-4240 spectrometer. Elemental analyses were performed by L Henling and F. Harvey at the Caltech Analytical Laboratory. Mass spectra were obtained from the UC Riverside facility. Cp*2 ScCI and Cp*2 ScCHa were prepared as described previously. 1 4 Cp*aScNHNH 2 (1 ). Method a: Cp*2 ScCI (2.54 g, 7.26 mmol) was dissolved in 40 mL of toluene. 5.3 mL of a 1.5M UCH 3 solution in diethyl ether (7.95 mmol) were added slowly by syringe to the solution, which was stirring under arqon. After 2 hr at room temperature, the volatiles were removed from the reaction mixture under reduced pressure. The residue was washed with 30 mL of petroleum ether and the resulting solution was filtered. Anhydrous hydrazine (22.5/tl, 7.1 mmol) was syringed into the solution with stirring under argon. The reaction mixture immediately turned from light-orange to maroon. After approximately twenty minutes the color dissipated and a white precipitate of 1 formed. The 166 precipitate was filtered and dried to yield 1.29g (51%) of product. IR: 3330,1565,1012, 712 cm*1. Anal. Calcd for C2 0 H3 3 N 2 SC: C, 69.4; H, 9.62; N, 8.09. Found: C, 67.83; H, 9.35, N, 8.08. MW Calcd: 345.9. MS largest peak: m /e = 346. Method b: An isolated sample of C p*2 ScCH3 was used. The reaction with hydrazine was performed under the same conditions as in method a. 0.837g of Cp*2 ScCH3 (2.54 mmol) afforded 0.518g of 1 (59%). C p * 2 ScNHN(CH 3 > 2 (2): Cp*2 ScCHs (1.0g, 3.0 mmol) was dissolved in 30 mL of petroleum ether. 1,1-Dimethylhydrazine, 3.0 mmol, were measured into a gas volume and condensed into the solution at -78 *C . The reaction mixture turned from light-orange to bright-yellow as it warmed to room temperature. After being stirred for one hour, the solution was concentrated and cooled to -78*C to afford yellow crystals of 2. Yield 0.45g (40%). IR: 3175, 2782, 2740,1208.1196,11 42.1080,1050,1016,995,850, 710, 650, 595, 550,409 cm*1. Anal. Calcd for C2 2 H 3 7 N 2 SC: C, 70.56; H, 9.96; N, 7.48. Found: C, 70.77; H, 9.57; N, 7.25. Cp#2 &cN(H)C(CH 3 )NNH 2 (3). Compound 1 (0.79g, 2.3mmol) was dissolved in 25 mL of methylene chloride. Acetonftrile, 0.5 mL, was vacuum transferred to the solution. The reaction was stirred for 30 min at room temperature. The volatiles were removed under vacuum and the white residue was taken up in 20 mL of petroleum ether as a slurry, which was cooled to -7B*C. The resulting white solid was filtered and dried. Yield 0.54g (61%). IR: 3370,3290,1607,1591,1540,1417,1355,1250,1230,1156,720, 550 cm'1. Anal. Calcd for C2 2 H3 sNsSc: C, 68.19; H, 9.36; N, 10.84. Found: C, 67.84; H, 9.10; N. 10.81. Cp*2 ScN(H)C(CH3 )riNMe2 (4). Compound 2 (0 . 1 45g, 0.388mmol) was dissolved in 15 mL of petroleum ether. Acetonitrile, 0.16 mL, was transferred to the solution. The reaction was stirred for 30 min at room temperature. Volatiles were removed under vacuum and the residue was dissolved in 3 mL of pentane and cooled at -78 *C overnight to 167 precipitate a tan powder, which was filtered cold and dried. Yield 0.059g (39%). IR: 3389, 1650,1506,1235,1033,1192,1033. Anal. Calcd for C2 4 H3 9 N4 SC: C, 69.4; H, 9.7; N, 10.1. Found: C, 69.3; H, 9.5; N, 9.75. Structure Determ ination fo r C p * 2 ScN(H)C(CH 3 >NNH2 . A colorless crystal obtained from an NMR tube reaction of Cp*2 ScNHNH 2 with CH 3 CN in deuterobenzene was mounted in a thin-walled glass capillary under N2 . The crystal was centered on a CAD-4 diffractometer. Unit cell parameters and an orientation matrix were obtained by a leastsquares calculation from the setting angles of 24 reflections. Two quadrants of intensity data out to a 28 of 40* and one quadrant of data in the 28 range 4 0 *-5 0 * were collected. No absorption or decay corrections were applied, and the data were reduced to F0 2 values and merged to yield the final data set. Systematic absences led to the choice of space group P2 i/n . The coordinates of the scandium atom were obtained from a Patterson map; locations of the other non-hydrogen atoms were determined from successive structure factor-Fourier calculations. Hydrogen atom positions for the methyl groups and for the heterometallacyclic ring were determined from difference maps and were refined along with their isotropic B values. The full least-squares matrix, consisting of coordinates, anisotropic Ufj's and isotropic B values for all of the atoms and a scale factor, contained 380 parameters. A final difference Fourier map showed deviations ranging from -0.81 eA"3 to O.BSeA"3. The refinement converged with an R factor of 0.117 (0.051 for F0 2 >3 <7 ) and a goodness of tit of 1.34 for all 3855 reflections. A summary of the data collection information is given in Table II, coordinates and Uaqs for non-hydrogen atoms are listed in Table III, parameters for the hydrogen atoms are listed in TAble IV, anisotropic displacement parameters are in Table V, and complete distances and angles are given in Table VI. Structure Determ ination for Cp*aScN(H)C(CH 3 )N N M e 2 . A colorless crystal grown from a petroleum ether solution of the complex cooled at -60*C was mounted in a 168 thin-walled glass capillary under N2 . The crystal was centered on a CAD-4 diffractometer. Unit-cell parameters and an orientation matrix were obtained by a least-squares calculation from the setting angles of 24 reflections with 35° <25 <39°. Two equivalent intensity data sets out to a 28 of 40° were collected, corrected for absorption (transmission coefficient: 0.889-0.956) and a slight decay, reduced to F0 2 values and merged to yield the final data set. Systematic absences led to the choice of space group P 2i/n. Coordinates of the scandium atom were obtained from a Patterson map; locations of the other non-hydrogen atoms were determined from successive structure factor-Fourier calculations. Hydrogen atom positions were determined from difference maps for the methyl groups and by calculation for the other one. All hydrogen atoms were given isotropic B values 20% greater than that of the attached atom; no hydrogen parameters were refined. The full least squares matrix, consisting of coordinates and anisotropic U ij' s for the non-hydrogen atoms and a scale factor, contained 253 parameters. A final difference Fourier map showed deviations ranging from -0.30 eA-3 to 0.38 eA*3. The refinement converged with an R factor of 0.0640 (0.0416 for F0 2 >3ct(F02)) and a goodness of fit of 1.78 for all 2267 reflections. A summary of the data collection information is given in Table VII, coordinates and Ueqs for non-hydrogen atoms are listed in Table VIII, parameters for the hydrogen atoms are listed in TAbie IX, anisotropic displacement parameters are in Table X, and complete distances and angles ate given in Table XI. All calculations were carried out on a VAX 11/750 computer using the CRYM system of programs and ORTEP. Scattering factors and corrections for anomalous scattering were taken from a standard reference (International Tables for X-ray Crystallography, Vol. IV, p. 71, p. 149; Birmingham, Kynoch Press, 1974). 169 References 1. St. Clair, M. A.; Santarsiero, B. 0.; Bercaw, J. E. Organometallics 1 9 8 9 ,8 , 17. 2. S t Clair, M. A. Ph.D. Thesis, California Institute of Technology, 1989. 3. Bercaw, J. E.; Davies, D. L; Wolczanski, P. T. Organometallics 1 9 8 8 ,5 ,4 4 3 . 4. Thompson, M. E. Ph.D. Thesis, California Institute of Technology, 1985. 5. Gagne, M. R.; Marks, T. J. J. Am. Cham. Soc. 1989,111,4108. 6. a) Leigh, G. J. In Recent Developments in Nitrogen Fixation Newton, W., Postgate, J. R., Rodriguez-Barruco, C., Eds.; Academic Press: New York, 1977; p. 1 . b) Chatt, J; Dilworth, J.R.; Richards, R .L Cham. Rev., 1978, 78(8), 589. c)Henderson, R A ; Leigh, G.S.; Pickett C.J. Adv. in Inorg. Cham, and Radiochem., 1983,2 7,197. c) 7. Stiefel, E. I. In Recent Developments in Nitrogen Fixation Newton, W., Postgate, J. R., Rodriguez-Barruco, C., Eds.; Academic Press: New York, 1977; p. 69. 8. a) Latham, IA ; Leigh G J. J. Chem. Soc. Dalton Trans., 1986,399. b) Clark, C.C.; Hydrazine; Mathieson Chemical Corporation: Maryland, 1953; Chapter 1. 9. a). McCleverty, J. A.; Rae, A. E.; Wolochowicz, I.; Bailey, NA ; Smith, J. M. A. J. Chem. Soc* Dalton. Trans., 1983,71. b) Murray, R. C.; Schrock, R. R. J. Am. Chem. Soc., 1 9 8 5 ,107,4557. c) Schrock, R. R.; Uu, A. H.; O'Regan, M. B.; Finch, W. C.; Payack, J. F. Inorg. Chem., 1988,27,3574. 10. a) Carroll, J. A.; Sutton, D; Cowie, M.; Gauthier, M. D. J. Chem. Soc* Chem. Common., 1979,1058. Carroll, J. A.; Sutton, D. Inorg. Chem., 1 9 8 0 ,1 9 ,3137. b) Cowie, Martin; Gautier, M. D. Inorg. Chem., 1 9 8 0 ,1 9 ,3142. c) C hatt, J.; Dilworth, J. R .; Dahlstrom, J. R.; Zubieta, P. L J. Chem. Soc* Chem. Common., 1980,786. d) Dilworth, J. R.; Latham, I. A., Leigh, G. J.; Huttner, G.; Jibril, I. J. Chem. Soc * Chem. Common., 1983,1368. e) Latham, I. A.; Leigh, G. J.; Huttner, G.; Jibril, I. J. Chem. Soc* Dalton Trans., 1986,385. 11. a) Barrientos-Penna, C. K.; Campana, C. F.; Einstein, F. W. B.; Jones. T.; Sutton, D.; Tracey, A. S. Inorg. Chem., 1 9 8 4 ,2 3 ,363. b) Nicholson, T.; Zubieta, J. A J. Chem. Soc* Chem. Common, 1985. c) Nicholson, T.; Zubieta, J. Polyhedron, 1 9 8 8 ,7 ,1 7 1 . 12. (a) FHzroy, M. d.; Frederiksen, J. M.; Murray, K. S.; Snow, M. R. Inorg. Chem., 1985, 2 4 ,3265. (b) Butcher, A. V.; Chatt, J.; Dilworth, J. R.; Leigh, G. J.; Hursthouse, M. B.; Jayaweera, S. A. A.; Quick, A. J. Chem. Soc* Dalton Trans., 1979,921. (c) Mattes, R.; Scholand, H.Angew. Chem. Int. Ed. Engl., 1983,22(3), 245. 13. a) Cotton, F. A.; Luck, R. L Inorg. Chem. 1 9 8 9 ,2 8 ,3210-3213. b) Brookhart, M.; Green, M. L H. J. Organomet. Chem. 1 9 8 3 ,2 SO, 395. 14. Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan, M. C.; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. J. Am. Chem. Soc. 1 9 8 7 ,109,203. 170 15. Pauling, L The Nature of the Chemical Bond; Cornell University Press: Ithaca, 1960; p. 228. 16. Other mechanisms may also be envisioned that would result the acetonitrile's being ultimately positioned C2C1-N2. For example, a mechanism analogous to that proposed for th e epoxidation of olefins by d° metal alkylperoxides (see Comprehensive Coordination Chemistry, Wilkinson, G.; Gillard, R. 0 .; McCleverty, J. A. Eds. Pergamon, Oxford, 1987; Volume 6 , Chapter 61.3, p 345), in which the nucleophilic olefin is replaced by nitrogen of incoming acetonitrile and the r;2-alkyl peroxide ligand with an »j2 -NHNH 2 . Following heterolytic cleavage o f the N-N bond, nucleophilic attack by the Sc-NI-fe at the carbon center would generate the first product in mechanism B (Scheme 1). 17. Collman, J. P.; Hegedus, L S.; Norton, J. R.; Finke, R. G. Principles end Applications of Organotransition Metal Chemistry. University Science Books: Mill Valley, CA, 1987; Chapter 7. 18. Hirabayashi T.; Itoh, K.; Sakai, S.; Ishii, Y. J. Organometal. Chem. 1970* 2 1,273. 19. Walsh, P. J.; Hollander, F. J.;Bergman, R. G. J. Am. Chem. Soc. 1 9 9 0 ,112,894. 20. Cohen, S. A.; Bercaw, J. E. Organometallics, 1 9 8 5 ,4,1006 21. Burger, B. J.; Bercaw, J. E. In New Developments in the Synthesis, Manipulation and Characterization of Organometallic Compounds; Wayda, A. L , Darensbourg, M. Y. Eds.; ACS Symposium Series 357; American Chemical Society: Washington, O. C., 1987; pp. 17-98. 22. Marvich, R. H.; Brintzinger, H. H. J. Am. Chem. Soc. 19 7 1 ,9 3 ,2046.